Re: Printing across multiple pages?
> Does anybody have a favorite tool for spitting a print job across > multiple pages (enlargement to be re-assembled / poor-man's large-format > output)? There's a 34"x94" poster at http://joannenova.com.au/2010/01/finally-the-new-revised-and-edited-climategate-timeline/ that comes complete or as a tape-it-together multipage .pdf. I think the latter was split by Acrobat (not Acrobat Reader). Not free, not open. I did print the first version on a color laser printer and trimmed & taped things together. Rather time consuming, but worked out better than I expected. I got a B&W version printed at The Copy Shop in Merrimack off Rt 101. $8.25, worked quite well. Kinko's estimated $125 for a color version. ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
Re: Why Linux has problems with proprietary multimedia...
Ralph, >So what it really comes down to is not who is in compliance with the >patent but who the patent-holders permit to violate their patent, how >well the patent holder was compensated to look the other way, and >whether at some future date the organization in this state of grace >manages to irritate the patent holder enough that the patent holder >goes after them anyway. >A near-perfect state of the eternal absence of real freedom for all >players involved, no matter how much they pay for it. >Dispensing injustice with consummate fairness. :) This is what happens when an idea is thought of as a piece of property and the patent is the recognition of the ownership of that property. If you own something, you can do anything you want with it. It is your right to sell a piece of property, or to burn it, or give it away. With patents you can even chose to prosecute some people and not others, as long as you can prove that it is not based on some bias that is illegal under the law. For example, it probably is not legal to prosecute only women based solely on their sex, or only blacks based solely on their skin color. The fact that the IP "belongs" to you whether or not you defend it is what allows "submarine" patents.patents which are not acted on until the patented item is in wide use, the person (perhaps innocently) violating the patent has invested millions in bringing out their product and the owner can get a lot of money in damages. Or, as in the case of Apple and HTC, the holder of the patent simply blocks them from doing business altogether. As to the "freedom" that is allowed or disallowed by patents, please refer to Article 1, Section 8 of the United States Constitution..the part that starts "To Promote the Progress of Science and useful Arts.."..it comes quite a bit before the "Bill of Rights": http://www.youtube.com/watch?v=oiyU2tiEB1g Warmest regards, md ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
Re: Android questions
On Fri, Mar 5, 2010 at 4:53 PM, Jerry Feldman wrote: > I currently have a Blackberry and will be moving to the Motorola > Backflip shortly. The only issue I have is that I have over 400 entries > in my Blackberry Memo Pad (currently synced with Outlook notes). There > are a number of ways that I can extract my memos from the Blackberry. > One objective is to be able to access the memo pad from Linux, which is > something I can't do with the Blackberry. > There is a program called ABC Amber that can take a Blackberry .ipd file > and convert it to several formats, including text, pdf, html, and other > formats. I do have some passwords stored there, but they will eventually > go into KeepPass. > > I've been using KeePassX. There is a Blackberry version which is a Java app. I bet there's an Android version. > There are many apps on Android that are similar to Blackberry's Memo > Pad. One program that would work fine is Evernotes, but I don't think > that I can access the Evernotes data directly from Linux, and I also > don't want to be locked into a single app. So the issue is going from > either the Blackberry or Windows Outlook Notes to the Android without > having to do a lot of work. Evernotes will work fine because it works > with Windows, Blackberry and Android, but AFAIK, not directly on Linux > platforms (though theoretically it might). > > So far, Evernote appears to be the best option because I can keep a copy > on Windows (work), Android, and on the web, so I can access the data > from my Linux system. So, what I'm really asking is to see what other > Android owners have done, primarily with Blackberry or Palm memo pads. > > Isn't there a web version of Evernote? There's always Google Docs too. ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
Re: Why Linux has problems with proprietary multimedia...
On 3/5/2010 6:27 PM, Joshua Judson Rosen wrote: > Joshua Judson Rosen writes: > > And, actually..., I saw came upon these interesting articles the other day: > > http://bemasc.net/wordpress/2010/02/02/no-you-cant-do-that-with-h264/ > > http://www.wedding-day-beauty.com/tag/final-cut > > > ... which reveal that even "Apple and Microsoft have paid up royalties..." > is only really about half-true: > So what it really comes down to is not who is in compliance with the patent but who the patent-holders permit to violate their patent, how well the patent holder was compensated to look the other way, and whether at some future date the organization in this state of grace manages to irritate the patent holder enough that the patent holder goes after them anyway. A near-perfect state of the eternal absence of real freedom for all players involved, no matter how much they pay for it. Dispensing injustice with consummate fairness. :) Ralph ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
Re: Why Linux has problems with proprietary multimedia...
Joshua Judson Rosen writes: > > "Jon 'maddog' Hall" writes: [...] > > H.264? Mpeg3/4/2? > > > > Have your friends send you Ogg Vorbis stuff. Plays fine. > > > > Apple and Microsoft have paid up royalties on these things (or at least > > Microsoft thought it had paid up royalties on mp3 until Alcatel/Lucent > > raised their hand a couple of years ago), so they can ship as many > > royalty-bearing codecs as they want. > > ... which has me wondering: how does Ubuntu get away with shipping all > of the stuff necessary to do DVD-authoring!? > > I looked into making DVDs with one of my Debian machines at one point, > and quickly accumulated a long list of things that had been intentionally > left out of Debian due to clear-and-present patent dangers, and that I > decided against pursuing *not* out of fear for the *technical* issues > involved (pshaw!) but out of fear that I end up setting myself up for > some patent-troll to `pursue a cross-licensing relationship with' me > (did I get that euphamism right?) in the future. And, actually..., I saw came upon these interesting articles the other day: http://bemasc.net/wordpress/2010/02/02/no-you-cant-do-that-with-h264/ http://www.wedding-day-beauty.com/tag/final-cut ... which reveal that even "Apple and Microsoft have paid up royalties..." is only really about half-true: While Apple and Microsoft have paid so that *they* can *distribute* the codecs, it turns out that they have *not* paid for *the users* to be able to *use them* in all cases. In the first article, Ben Schwartz explores the prohibitive language used in the EULA for `Final Cut Pro', ultimately deciding that "Final Cut Hobbyist" would be less misleading. And after looking at the the EULA for Windows 7 Ultimate, he says: "Doesn't seem so Ultimate to me.". In the second article, the (different) author actually follows-up and asks the various involved parties (Apple, Microsoft, MPEG LA): "Schwartz's contention caught my attention: my SLR shoots 1080p video encoded with H.264, and I'm in a position both to publish some videos online for my main job and sell others on the side. and with bubbling controversies regarding how HTML is reshaping online video, any troubles with H.264 constraints take on new interest. "It seemed like a good time to call Apple, Adobe, and the MPEG-LA, the industry group that licenses the H.264 patent portfolio to the likes of software companies, optical-disc duplicators, Blu-ray player makers, and others who have need to use H.264." After a bit of run-around, it seems, the determination is made that yes, many individuals *do* need to pay their own added licensing-fees, for use of legitimately-obtained the codecs: "When I heard back from Allen Harkness, MPEG LA's director of global licensing, though, I was relieved to learn that Final Cut Pro isn't just for making YouTube cat videos. "But H.264 use isn't all free all the time--the wedding videographer might need to pay 2 cents per disc they sell, for example--and even experts can be thrown off by the complications." It's worth reading both of these articles in their entirety. Ben Schwartz ends, I gather, in agreement with maddog: "My advice: use a codec that doesn’t need a license: _Q. What is the license for Theora?_ Theora (and all associated technologies released by the Xiph.org Foundation) is released to the public via a BSD-style license. It is completely free for commercial or noncommercial use. That means that commercial developers may independently write Theora software which is compatible with the specification for no charge and without restrictions of any kind." Me too. -- "Don't be afraid to ask (λf.((λx.xx) (λr.f(rr." ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
Re: Why Linux has problems with proprietary multimedia... (was: Interesting article)
>... which has me wondering: how does Ubuntu get away with shipping all >of the stuff necessary to do DVD-authoring!? Ahhh, what does it meant to do "DVD-authoring"? Moving encoded bits on a DVD? No problem! Taking video bits from my video camera (encoded into Mpeg) and putting it onto my DVD? No problem! Making a DVD of Ogg Theora? No problem! Encoding? Depends on the patents involved, the licensing around the patents, and so forth. The results of the encoding? The H.264 patent group has recently released yet another wave of grace over mpeg-4 streams "free to end users" would not have to have royalties paid on the *streams*. Of course what "free to end users" is creates another whole bag of worms. >I looked into making DVDs with one of my Debian machines at one point, >and quickly accumulated a long list of things that had been >intentionally >left out of Debian due to clear-and-present patent dangers, and that I >decided against pursuing *not* out of fear for the *technical* issues >involved (pshaw!) but out of fear that I end up setting myself up for >some patent-troll to `pursue a cross-licensing relationship with' me >(did I get that euphamism right?) in the future. You have to watch those relationships with Trolls.they create really ugly offspring. Now I think I am going to have a beer.I need a beer md ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
Why Linux has problems with proprietary multimedia... (was: Interesting article)
"Jon 'maddog' Hall" writes: > > > Not one Linux distro I've seen does a convincing job with consumer > > media, an absolutely basic requirement, and something we ought to be > > able to get right. > > Well, please ask the DVD people not to used royalty bearing patents in > their codecs, and encryption practices that would have the DMCA down on > the headquarters of Fedora, OpenSUSE and others. > > > movie or sound file > > H.264? Mpeg3/4/2? > > Have your friends send you Ogg Vorbis stuff. Plays fine. > > Apple and Microsoft have paid up royalties on these things (or at least > Microsoft thought it had paid up royalties on mp3 until Alcatel/Lucent > raised their hand a couple of years ago), so they can ship as many > royalty-bearing codecs as they want. ... which has me wondering: how does Ubuntu get away with shipping all of the stuff necessary to do DVD-authoring!? I looked into making DVDs with one of my Debian machines at one point, and quickly accumulated a long list of things that had been intentionally left out of Debian due to clear-and-present patent dangers, and that I decided against pursuing *not* out of fear for the *technical* issues involved (pshaw!) but out of fear that I end up setting myself up for some patent-troll to `pursue a cross-licensing relationship with' me (did I get that euphamism right?) in the future. -- "Don't be afraid to ask (λf.((λx.xx) (λr.f(rr." ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
Re: Interesting article
Ralph, While I agree with some things you said: >Not one Linux distro I've seen does a convincing job with consumer >media, an absolutely basic requirement, and something we ought to be >able to get right. Well, please ask the DVD people not to used royalty bearing patents in their codecs, and encryption practices that would have the DMCA down on the headquarters of Fedora, OpenSUSE and others. >movie or sound file H.264? Mpeg3/4/2? Have your friends send you Ogg Vorbis stuff. Plays fine. Apple and Microsoft have paid up royalties on these things (or at least Microsoft thought it had paid up royalties on mp3 until Alcatel/Lucent raised their hand a couple of years ago), so they can ship as many royalty-bearing codecs as they want. I was looking into what it would take in the way of patent royalties to put Android onto the Openmoko phone. It was a mess, even just paying the royalties on a hardware basis. But people can not afford to pay the royalties on free CDs that they give away and may never even install...or royalties for downloads that may not even make it to CDs. Sooo, installing these codecs is a "research project" as you call it. >However, if Linux were an attractive entertainment platform in all >other ways, I suspect people would put it on their computers for that - >can't beat the price - and then companies would start writing games for >the platform. Until you can at least play a DVD after first boot, >forget it. Too much trouble. Not even close. There were a couple of projects that did this. LinuxMCE comes to mind. There were even companies that had products based on it. Hmmm, LinuxMCE seems to have gotten a new lease on life. I may try it again. md ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
Android questions
I currently have a Blackberry and will be moving to the Motorola Backflip shortly. The only issue I have is that I have over 400 entries in my Blackberry Memo Pad (currently synced with Outlook notes). There are a number of ways that I can extract my memos from the Blackberry. One objective is to be able to access the memo pad from Linux, which is something I can't do with the Blackberry. There is a program called ABC Amber that can take a Blackberry .ipd file and convert it to several formats, including text, pdf, html, and other formats. I do have some passwords stored there, but they will eventually go into KeepPass. There are many apps on Android that are similar to Blackberry's Memo Pad. One program that would work fine is Evernotes, but I don't think that I can access the Evernotes data directly from Linux, and I also don't want to be locked into a single app. So the issue is going from either the Blackberry or Windows Outlook Notes to the Android without having to do a lot of work. Evernotes will work fine because it works with Windows, Blackberry and Android, but AFAIK, not directly on Linux platforms (though theoretically it might). So far, Evernote appears to be the best option because I can keep a copy on Windows (work), Android, and on the web, so I can access the data from my Linux system. So, what I'm really asking is to see what other Android owners have done, primarily with Blackberry or Palm memo pads. -- Jerry Feldman Boston Linux and Unix PGP key id: 537C5846 PGP Key fingerprint: 3D1B 8377 A3C0 A5F2 ECBB CA3B 4607 4319 537C 5846 signature.asc Description: OpenPGP digital signature ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
Re: Linux for "cloud computing": Request for Input
Also Process Software's MultiNET, which we were using circa '98-2000 at one site here in Vermont. On Fri, Mar 5, 2010 at 4:39 PM, Jon 'maddog' Hall wrote: > Mark, > > >There is still some good stuff happening with VMS, for example if you > >are an hp software partner, you can get ssh access to a virtual > >machine running OpenVMS 8.4 EFT. > > I did not mean to imply that there was not "good stuff happening with > VMS"but VMS is and was not DECnet. > > Even when I was there you could see the writing on the wall for DECnet > as a protocol. TCP/IP was available for VMS, first through Wollongong > (boy, I have not thought about them in years!) and then through DEC > itself with TCP/IP Services for VMS (then TCP/IP Services for OpenVMS, > now HP TCP/IP Services for OpenVMS). > > BUT people still had LAT boxes, and wanted to use them. And LAT boxes > did not speak TCP/IP. So there were gateway products and such created, > both by DEC and then by others. But the gateway products just were not > the same as "DECnet". > > DECnet Linux was just another fine example of FOSS extending the life of > otherwise forgotten hardware. > > I went to the HP site and found the latest version of DECnet OpenVMS > (Version 7.3). The date on the manual was May of 1993, one year before > I met Linus. > > DECnet-Plus for OpenVMS (including X.25 support) latest version is 8.3, > with a date of June 2006. > > RIP DECnet Linux! > > md > > ___ > gnhlug-discuss mailing list > gnhlug-discuss@mail.gnhlug.org > http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/ > ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
Re: Linux for "cloud computing": Request for Input
Mark, >There is still some good stuff happening with VMS, for example if you >are an hp software partner, you can get ssh access to a virtual >machine running OpenVMS 8.4 EFT. I did not mean to imply that there was not "good stuff happening with VMS"but VMS is and was not DECnet. Even when I was there you could see the writing on the wall for DECnet as a protocol. TCP/IP was available for VMS, first through Wollongong (boy, I have not thought about them in years!) and then through DEC itself with TCP/IP Services for VMS (then TCP/IP Services for OpenVMS, now HP TCP/IP Services for OpenVMS). BUT people still had LAT boxes, and wanted to use them. And LAT boxes did not speak TCP/IP. So there were gateway products and such created, both by DEC and then by others. But the gateway products just were not the same as "DECnet". DECnet Linux was just another fine example of FOSS extending the life of otherwise forgotten hardware. I went to the HP site and found the latest version of DECnet OpenVMS (Version 7.3). The date on the manual was May of 1993, one year before I met Linus. DECnet-Plus for OpenVMS (including X.25 support) latest version is 8.3, with a date of June 2006. RIP DECnet Linux! md ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
Re: Interesting article
On Fri, Mar 5, 2010 at 10:41 AM, Alan Johnson wrote: >>> > Well, except MacOSX has specific hardware. >>> >>> >>> >>>Indeed, that's a big part of Apple's strategy. Design the hardware >>> >>> and the software together, and they'll work well together. And there >>> >>> is something to be said for that. >>> >> >> >> >> And that's one of Apple's prime advantages. >> > > > > Advantage? Well, for Apple, it is an advantages over MS, but certainly not > > for the users. > Those of us on this list are passionate about computers. But let's face it, unless you're a geek, a computer is about as interesting as a telephone or a toaster oven or a DVD player. In the real world, a computer is a convenient appliance for interpersonal communication and entertainment. It shouldn't be something you have to spend time on. If you want to make Linux work for the world, you need to provide an out-of- the-box experience where you install the OS and it JUST WORKS with all the obvious tools close at hand. Apple and at least the better Windows box vendors have managed to do a decent job at this. Linux hasn't until recently. Ubuntu seems to provide a good basic level of first-boot usability in most things. I've had no problem with sound or wireless on my HP laptop running Ubuntu 9.04, common areas of difficulty on some distros I've tried. I have had some odd issues with repaints after the system has been idle for a while, though. (I'm not sure if this is an issue with the NVidia drivers.) That kind of thing wouldn't be appropriate in an appliance. Not one Linux distro I've seen does a convincing job with consumer media, an absolutely basic requirement, and something we ought to be able to get right. I should be able to use my computer to do anything that the gear in my living room can do without having to expend effort to make it work - play DVDs using the menus, use webcams like vidcams, record television like a DVR. If a friend sends you a movie or sound file, you should be able to just click it and play it without it turning into a research project. We'll get there but we aren't there yet. Once you can do that on Linux, you have a potential consumer market - those who don't play computer games. :) The one piece of living room gear that may prove an intractable challenge for Linux to match is the game console. I have always had to keep a Windows system around for games and some Windows-only applications. I haven't yet found a package on Linux that actually emulates Windows well enough to play games without endless experimentation. I haven't found one useful Windows XP program for which Wine works and during the period I subscribed to Cedega, somebody had made just about every game work, it seemed, except the few I made time to play. Go figure. No surprises there, though. Being Windows reliably is an intractable problem. Even Microsoft has a hard time pulling it off. :) However, if Linux were an attractive entertainment platform in all other ways, I suspect people would put it on their computers for that - can't beat the price - and then companies would start writing games for the platform. Until you can at least play a DVD after first boot, forget it. Too much trouble. Not even close. Lupestro ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
Re: Printing across multiple pages?
http://lifehacker.com/233563/weekend-project-create-a-poster-from-any-image-with-block-posters http://lifehacker.com/5468489/easy-poster-printer-slices-and-dices-your-posters-for-standard-printers On Fri, Mar 5, 2010 at 4:02 PM, Bruce Dawson wrote: > I assume 'pdftops foo.pdf | pstopnm | pnmscale x | gv -' makes things > too grainy. > > Have you tried scribus or inkscape? I believe one of them (but I forget > which) will manipulate .pdf files. > > --Bruce > > Bill McGonigle wrote: > > Does anybody have a favorite tool for spitting a print job across > > multiple pages (enlargement to be re-assembled / poor-man's large-format > > output)? > > > > I've tried a few tools which claim to do that, but they've all failed to > > take a run-of-the-mill PDF chart and turn out working output. > > > > Last time I tried I wound up using screen-grabs and GIMP, and that was > > really a dumb way to approach it. > > > > Thanks, > > -Bill > > > > > > ___ > gnhlug-discuss mailing list > gnhlug-discuss@mail.gnhlug.org > http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/ > ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
Re: Linux for "cloud computing": Request for Input
On Fri, Mar 5, 2010 at 3:35 PM, Jon 'maddog' Hall wrote: > So just like I said "RIP" to Grace Murray Hopper, I now say "RIP" to > DECnet Linux. There is still some good stuff happening with VMS, for example if you are an hp software partner, you can get ssh access to a virtual machine running OpenVMS 8.4 EFT. See: http://bit.ly/bmNyDE Or start at http://www.hp.com/go/dspp and drill down a bit to the Partner Virtualization Program part. -marc -- Marc Nozell (m...@nozell.com) http://www.nozell.com/blog ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
Re: Printing across multiple pages?
I assume 'pdftops foo.pdf | pstopnm | pnmscale x | gv -' makes things too grainy. Have you tried scribus or inkscape? I believe one of them (but I forget which) will manipulate .pdf files. --Bruce Bill McGonigle wrote: > Does anybody have a favorite tool for spitting a print job across > multiple pages (enlargement to be re-assembled / poor-man's large-format > output)? > > I've tried a few tools which claim to do that, but they've all failed to > take a run-of-the-mill PDF chart and turn out working output. > > Last time I tried I wound up using screen-grabs and GIMP, and that was > really a dumb way to approach it. > > Thanks, > -Bill > > ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
Re: Linux for "cloud computing": Request for Input
On Fri, Mar 5, 2010 at 3:12 PM, Jon 'maddog' Hall wrote: > Bill, > > >ZFS is only on *Solaris and FreeBSD (albeit an old one). Linux > >doesn't yet have a stable, consistent COW filesystem. Certainly a > >combination of the two is a great win. > > It has been some time since I have looked at file systems, and > particularly COW file systems, so pardon me if these questions are > "naive". > COW makes snapshotting easy and can reduce the need for a fsck. NetApp's WAFL is a COW. With WAFL or ZFS, you can have snapshots every 15 minutes and keep multiple copies of the snapshot w/o using significant disk space. BSD (sun's UFS) and LVM allow only 1 snapshot at a time. For backup of a database: freeze the DB, snapshot, start the DB, backup the snapshot, delete the snapshot. Any traction to ext3cow or using the COW layering capability of the UDB > block driver? Or LVM "snapshots"? > > And I can not remember if GFS (Global File System, or even the Google > File System) was COW. > I think GFS is a shared file system. Over iSCSI, Fibre Channel or Infiniband. It can be much faster then NFS and reduces the bottleneck. > Finally, what about the up and coming Btrfs? > >From what I've read it looks like a nimby ZFS with improvements. As has been mentioned, Linux reinvents things instead of building on others sometimes. In this case, because the ZFS license isn't compatible with the GPL in the Linux kernel. And that's a valid reason. OpenSSH was created for similar reasons. btrfs changed the underlying code that should make it much easier to reduce a pool to fewer or smaller disks or increase a RAID5 by adding another disk like some RAID cards allow. Some of the ZFS discussions assume that people will just build another pool and only home user types would do this. It's kindof like seening 2 mice; one is a marsupial and one is a mammal. They're built very differently but function similarly. I'm hoping btrfs takes off and becomes part of Linux as a viable ZFS alternative. ReiserFS is a good example. The transition from 3 to 4 hurt its inclusion in the kernel. Does SuSE still use it as the default FS? Are people still using it? There are lots of projects on filesystems these days. SSDs, Embedded systems, Distributed nodes, HPC, reliability. It's a good thing. ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
Re: Linux for "cloud computing": Request for Input
Tom, >Networking Linux can do vlans, VPN, firewalls in the base install. Its >very flexible in what you allow to be exposed. AWESOME with the networking mention. The BSD guys often said in the early days that they had a better networking stack, but I am fairly sure that Linux has caught up. :-) More than that, Linux supported more networking types than any other OS I know. X.25, uucp...you name it. Linux even has a FOSS version of DECnet. /* Aside The project leader for the DECnet Linux project wrote to me after DEC had been purchased by Compaq and Compaq was "merged" with HP, and asked if I thought that "DEC" would mind the project using the name "DECnet". I gave him the "Rear Admiral Grace Murray Hopper" answer: "It is easier to ask forgiveness than permission." and in this case it was literally true, as I did not know if there was anyone left to the DECnet group at HP. The DECnet Linux project gave DECnet Phase IV functionality to Linux, which allowed Linux to talk to VMS systems that only had DECnet, and to LAT boxes (this was a bunch of serial lines tied into a box and then hooked to the ETHERNET) which would handle terminals, printers and other serial devices. I thought that this project added to DECnet's life, and that DEC/Compaq/HP would have to be crazy to object. But sadly I now see (as of February 10th, 2010): http://www.csamuel.org/2010/02/19/decnet-now-orphaned-in-the-linux-kernel-for-2-6-33 So just like I said "RIP" to Grace Murray Hopper, I now say "RIP" to DECnet Linux. */ Warmest regards, maddog ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
Re: Linux for "cloud computing": Request for Input
On Fri, Mar 5, 2010 at 2:58 PM, Bill McGonigle wrote: > On 03/05/2010 02:45 PM, Tom Buskey wrote: > >> Samba works well here. ZFS also has a CIFS server built in that does >> all the ACLs that Windows needs. >> > > But you still have to boot Windows off of a "C:" block-device, right? (and > run your many apps that only run on C:) If there's a CFS-C:\ like Linux > NFS-root that would be really helpful. That way you could go into, e.g., > C:\WINDOWS\SYSTEM32\.zfs\ to fix your brokenness. :) > A fileserver won't help with apps that want everything on C:. It will only work for D: I'm not sure Windows can boot off an iSCSI (or FC) target either. Solaris can. I don't know if Linux can. Then there are network boots. Linux and Solaris excel at this. Very useful in provisioning. > > I haven't been able to find this feature, but I also don't really know what > to search for. > > I've always thought that propagating C:, D:, PRN:, etc was one of the poorer things that NT kept from DOS (and is predecessors). It's useful to slip a 2nd drive into a system and move /var to it when space is needed. Or move /usr/local to a fileserver when the workstations have small (400 MB) drives. (DON'T do this. rpm won't like you and you'll create other headaches) ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
Re: Linux for 'cloud computing': Request for Input
On Fri, March 5, 2010 3:12 pm, Jon 'maddog' Hall wrote: > It has been some time since I have looked at file systems, and > particularly COW file systems, so pardon me if these questions are > "naive". > > Any traction to ext3cow or using the COW layering capability of the UDB > block driver? Or LVM "snapshots"? I'd say non-btrfs COW solutions are being back-burnered. When Ted T'so (for example), primary guy behind ext-4, says it's a stopgap until btrfs, you get the idea that folks are hot for it. And with people like Valerie Aurora (a former ZFS developer) watching closely and offering advice, you hope that they get things right. > And I can not remember if GFS (Global File System, or even the Google > File System) was COW. Dunno. > Finally, what about the up and coming Btrfs? See above. It's coming. I'd say that, by fall, you could start using it as your primary, non-server filesystem. Since folks are (understandably) very conservative about filesystems, it'll probably be another two years before it's a primary install option, but it's coming. It's got some truly neat features: - COW - Explicit/online defragmentation (volume or file) - Filesystem-aware RAID (you can even alter RAID on a per-file basis) - VERY flexible, filesystem-aware snapshotting - Online fsck - Checksums (yay!) - (File-level) de-duplication; block-level is being discussed, but apparently would require an on-disk format change, so that's probably a major rev away. (Ken's guess.) - Etc. See the Wikipedia page or the btrfs Wiki for more info. (http://en.wikipedia.org/wiki/Btrfs and http://btrfs.wiki.kernel.org , respectively.) -Ken -- This message has been scanned for viruses and dangerous content by MailScanner, and is believed to be clean. ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
Re: Printing across multiple pages?
On Fri, Mar 05, 2010 at 03:18:35PM -0500, Bill McGonigle wrote: > Does anybody have a favorite tool for spitting a print job across > multiple pages (enlargement to be re-assembled / poor-man's large-format > output)? > > I've tried a few tools which claim to do that, but they've all failed to > take a run-of-the-mill PDF chart and turn out working output. I've used the online service here before: http://homokaasu.org/rasterbator/ It worked reasonably well. -ben -- this is an unbearable strain, but i'm doing it as hard as i've ever done it before. ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
Printing across multiple pages?
Does anybody have a favorite tool for spitting a print job across multiple pages (enlargement to be re-assembled / poor-man's large-format output)? I've tried a few tools which claim to do that, but they've all failed to take a run-of-the-mill PDF chart and turn out working output. Last time I tried I wound up using screen-grabs and GIMP, and that was really a dumb way to approach it. Thanks, -Bill -- Bill McGonigle, Owner BFC Computing, LLC http://bfccomputing.com/ Telephone: +1.603.448.4440 Email, IM, VOIP: b...@bfccomputing.com VCard: http://bfccomputing.com/vcard/bill.vcf Social networks: bill_mcgonigle/bill.mcgonigle ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
Re: Linux for "cloud computing": Request for Input
Bill, >ZFS is only on *Solaris and FreeBSD (albeit an old one). Linux >doesn't yet have a stable, consistent COW filesystem. Certainly a >combination of the two is a great win. It has been some time since I have looked at file systems, and particularly COW file systems, so pardon me if these questions are "naive". Any traction to ext3cow or using the COW layering capability of the UDB block driver? Or LVM "snapshots"? And I can not remember if GFS (Global File System, or even the Google File System) was COW. Finally, what about the up and coming Btrfs? Thanks, md ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
Re: Linux for "cloud computing": Request for Input
On 03/05/2010 02:45 PM, Tom Buskey wrote: > Samba works well here. ZFS also has a CIFS server built in that does > all the ACLs that Windows needs. But you still have to boot Windows off of a "C:" block-device, right? (and run your many apps that only run on C:) If there's a CFS-C:\ like Linux NFS-root that would be really helpful. That way you could go into, e.g., C:\WINDOWS\SYSTEM32\.zfs\ to fix your brokenness. :) I haven't been able to find this feature, but I also don't really know what to search for. -Bill -- Bill McGonigle, Owner BFC Computing, LLC http://bfccomputing.com/ Telephone: +1.603.448.4440 Email, IM, VOIP: b...@bfccomputing.com VCard: http://bfccomputing.com/vcard/bill.vcf Social networks: bill_mcgonigle/bill.mcgonigle ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
Re: Linux for "cloud computing": Request for Input
>Virtualization? Well, I did not mention virtualization before, since there are so many answers for virtualization for different operating systems, but there is an element of efficiency in virtualization and >Too many choices. A good point, but I think it is overblown for "uber" cloud computing (Google, Amazon, Oracle, and other large "service" vendors) who either roll their own (because they are big enough, and they can) or partner with one of the "big guys" (Red Hat, Novell). More of an issue for "private clouds", but I think that enterprise companies will still go with "Enterprise vendors" like Red Hat, Novell, Canonical, if only that their "private cloud" will match with their "public cloud". Small companies and ISPs that offer "cloud services" may see more choice indecision, but they will have people who can analyze which is best or will hire consultants, and that is why they will be paid the big bucks. :-) I do not think the "indecision factor" will be as great for cloud computing as for the noobie trying to choose their first desktop distro. Good points though. md ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
Re: Interesting article, games
On 03/05/2010 12:36 PM, Thomas Charron wrote: > But there's a whole slew of > custom-google-glue-and-tweaks, which was disappointing. Yeah, Google likes to grab a project, modify it to its own needs, and throw the code over the wall. Same as happens for Chrome. Tom Callaway at Fedora did yeoman's work figuring this all out, building patch files, proper RPM's, etc., but then found that there were licensing violations too and had to abandon it. Google could do this work if it wanted to (it was a part-time effort for Tom) but they choose not to. The Android stuff has been removed from linux for now since their sleep locking stuff was only optimized for time-to-market. The community-friendly project that's most like Android is MeeGo, the merger of Maemo (Nokia) and Moblin (Intel), which is now essentially a Fedora derivative. It would be lovely if Google participated in the community too. -Bill -- Bill McGonigle, Owner BFC Computing, LLC http://bfccomputing.com/ Telephone: +1.603.448.4440 Email, IM, VOIP: b...@bfccomputing.com VCard: http://bfccomputing.com/vcard/bill.vcf Social networks: bill_mcgonigle/bill.mcgonigle ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
Re: Linux for "cloud computing": Request for Input
On Fri, Mar 5, 2010 at 2:28 PM, Bill McGonigle wrote: > On 03/05/2010 02:03 PM, Brian St. Pierre wrote: > > Virtualization? > > Yes, and to drive home that point, this is what's being chosen > empirically by the extant service providers. Amazon's EC2 is Xen on > RHEL. I seem to recall that Rackspace also went this route. > > And if you don't like Xen on RHEL/CentOS you can switch to KVM on RHEL. > Or (heaven forbid) VMWare Server on RHEL or VMWare ESX which was > partially based on RHEL. Even Xen started life with a big code import > from Linux. > > And if you go the Xen route and decide you really don't like Linux you > can run Xen on OpenSolaris. Open Source is good - you can choose Linux > first and still have good escape routes. > Hey, I was going to say that. VirtualBox is another interesting one. They demoed moving a running guest VM (running Solaris I think) from a MacOSX host to a Windows 7 host while running. Virtualization means you don't have to reinstall from OEM disks when your upgrade the server hardware. Just down the VM, copy the images to the new server, and start it again. To the VM running on the host, the network, storage controller, display, etc don't change. I think Linux is the best host for VirtualBox right now. Solaris is probably the least supported. Networking Linux can do vlans, VPN, firewalls in the base install. Its very flexible in what you allow to be exposed. > > >> > Any blatant negatives for Linux as a platform? > > ZFS is only on *Solaris and FreeBSD (albeit an old one). Linux doesn't > yet have a stable, consistent COW filesystem. Certainly a combination > of the two is a great win. > > There is FUSE, but I wouldn't use it if I had any choice at all. > If you assume ZFS for storage virtualization, Linux has the advantage of > being able to (p)NFS mount the data, so you can do file-level > snapshotting and data de-duplication. You could also run Windows-based > clouds with a ZFS backend, but you'd have to use an iSCSI backend which > loses the nice snapshotting capabilities and drives everything back to > the block device level of granularity because Windows doesn't play > nicely with everybody else. > Samba works well here. ZFS also has a CIFS server built in that does all the ACLs that Windows needs. iSCSI for any database where you can't use a file server. ZFS does deduplication in the pools. Which means it applies to iSCSI too. Only on OpenSolaris development b131 and higher or the Sun Storage appliances. ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
Re: Linux for "cloud computing": Request for Input
On Fri, Mar 5, 2010 at 12:46 PM, Jon 'maddog' Hall wrote: > Hi, > > I am doing some consulting about "Why Linux is good for cloud > computing" (and for that matter "Software as a Service" (SaaS), which I > consider more or less one and the same). > > I am going to start with the fact that Unix systems were designed > (almost) from the beginning to be multi-tasking and multi-user, then > worked its way out to be multi-architecture, multi-threaded and embraced > 64-bit systems relatively early. > > In addition, Unix systems utilized network processes and stressed > client/server models (evident in daemons, NFS, the X Window System), as > well as scripting languages that allowed control of applications through > APIs and not just a graphical interface. > > This gave: > > o scalability > o some architectural security > o standards based development > > A lot of work in highly available servers was done with Unix systems > such as Solaris, Digital Unix and others. > > Linux, patterned after Unix, inherited a lot of these characteristics. > > In addition with Linux you get: > > o "Open Source" that helps give: > > - needed bug fixes rapidly (under control of the service provider) > - development of new features by large numbers of programmers, > researchers, etc. > - many "middleware" and emerging "management" systems are being > developed on Linux, or developed as FOSS projects > + Eucalyptus > + Languages like Ruby, PHP, Python > - licensing terms that do not restrict what you can offer to > customers (i.e. how many instances can you run, how many customers can > attach, etc.) > - a couple of different security models (SE/Linux, AppArmor, as > examples) to chose from > - graceful degradation: If a technology is abandoned, the provider > of services can maintain it until a migration can occur through > community action > - Open development model - allows service providers to plan ahead and > have input to development > > o Linux also was used as the basis for Beowulf systems, which developed > a lot of code surrounding "high performance clusters", leading to highly > scalable systems > > o Basically can be same OS on desktop as servers > > Any other ideas on the topic of "Why Linux for Cloud Computing?" > Right sizing of the OS. 1) Space: You can chose exactly what goes into the OS image on the cloud. No browser, no wordpad.exe taking up space. 2) Function: Only install what is needed. If notepad is the editing app, that what is installed and nothing else needs to be maintained, educated, patched. 3) Security: If you don't install acrobat, your server isn't affected by any security holes in it. Application availability: 1) Many web apps/frameworks are developed on Linux 1st. Then they get ported to other Unixen. > > Any blatant negatives for Linux as a platform? > > Note that I am not arguing for or against "the cloud", just why Linux is > or is not a good system for it. > > The entire paper will be available for free download when it is > finished. > > Thanks in advance for your input. > > md > > ___ > gnhlug-discuss mailing list > gnhlug-discuss@mail.gnhlug.org > http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/ > ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
Re: Linux for "cloud computing": Request for Input
On 03/05/2010 02:03 PM, Brian St. Pierre wrote: > Virtualization? Yes, and to drive home that point, this is what's being chosen empirically by the extant service providers. Amazon's EC2 is Xen on RHEL. I seem to recall that Rackspace also went this route. And if you don't like Xen on RHEL/CentOS you can switch to KVM on RHEL. Or (heaven forbid) VMWare Server on RHEL or VMWare ESX which was partially based on RHEL. Even Xen started life with a big code import from Linux. And if you go the Xen route and decide you really don't like Linux you can run Xen on OpenSolaris. Open Source is good - you can choose Linux first and still have good escape routes. >> > Any blatant negatives for Linux as a platform? ZFS is only on *Solaris and FreeBSD (albeit an old one). Linux doesn't yet have a stable, consistent COW filesystem. Certainly a combination of the two is a great win. If you assume ZFS for storage virtualization, Linux has the advantage of being able to (p)NFS mount the data, so you can do file-level snapshotting and data de-duplication. You could also run Windows-based clouds with a ZFS backend, but you'd have to use an iSCSI backend which loses the nice snapshotting capabilities and drives everything back to the block device level of granularity because Windows doesn't play nicely with everybody else. -Bill -- Bill McGonigle, Owner BFC Computing, LLC http://bfccomputing.com/ Telephone: +1.603.448.4440 Email, IM, VOIP: b...@bfccomputing.com VCard: http://bfccomputing.com/vcard/bill.vcf Social networks: bill_mcgonigle/bill.mcgonigle ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
Re: Interesting article,
Between everyone here I continue to learn a helluva lot about what's going on with Linux vs. everything else, and am always grateful for it, particularly for the input from md and Ben recently. So, while having nothing much to contribute at this time other than congratulations and thanks for such intel and opinion, my best wishes from northern Vermont as the snow begins? to melt away and the sugar sap is running steady. Old Farmer Davy Pavilion Farm (1806) West Montpeculiar VAX/VMS and OpenVMS ex-operator, sys admin, and "infrastructure analyst" and Alpha fan (hey, remember the AlphaServers? running VMS, NT or Red Hat while using practically zip for RAM or disk space???) (engineers, developers and sys admins went first to NT then to India then to Linux) DEC and HP-UNIX sys admin WinNT sys admin Windows Servers sys admin Red Hat sys admin Unemployed (until md needs a bodyguard and/or door gunner for his travels worldwide) On Fri, Mar 5, 2010 at 2:10 PM, Jon 'maddog' Hall wrote: > O.K., I will wade in here. :-) > > For the most part, Ben is right. Vendors who completely control both > hardware and software can make the "best" products, if your definition > of "best" is a limited market of items, and you are willing to pay for > them. MVS, VMS, Digital Unix. Rock solid, stable, scalable. REALLY > EXPENSIVE. > > A lot of Microsoft's problems have to do with drivers that come from > different vendors, trying to control different controllers that fit into > buses that are not that well documented. One mistake in a driver > (inside the monolithic kernel) and BAM! Lockup and blue screen. I am > amazed that Microsoft's eco-system can actually boot at all. > > >> Apple could have > >> crushed MS by now if they had gone with the GPL attitude instead of > >> picking BSD so they could keep all their toys to themselves. > > >Yah, I'm not buying that. If all you needed was the GPL, Linux > >would already have crushed Microsoft. > > Ben, I think you are under-estimating your own argument about inertia. > > Inertia is all about acceleration, not really speed. In 1991 Microsoft > was already going 50,000 mph and accelerating and Linux started from > zero, with almost zero acceleration. In 1994 a lot of the vendors > seemed as if they were going to give the server market to WNT. There > were a lot of people saying that "Unix was dead". > > After twenty years Microsoft is still accelerating, but I think it is > accelerating at a slower pace, and FOSS is accelerating at a faster > pace, but has still not caught up. Then there is distance traveled, or > "speed over time" (in this case, installed base). It may take a very, > very long time before FOSS has the same installed base, much less > "crushing Microsoft". > > Apple has existed for about the same time as Microsoft, and still has > about the same market penetration as twenty years ago. Its acceleration > is a lot slower, and more or less allowed it to keep the same desktop > and server market share (or maybe lost server market share in that > time). > > >(When Linux first came out, you also had a fleet of commercial Unixes, > >Novel, several BSDs, OS/2, BeOS, and all sorts of other bit player > >platforms. Today it's all Microsoft, with Apple and Linux nipping at > >their heels. > > Sadly, and from a "choice" and "research" viewpoint this is true. But > the fragmentation meant that unless any of them reached critical stage, > they would be just what you said "bit players", and would have died > anyway. > > > Absolutely correct! However, the fact that's it's a hard problem to > >solve doesn't mean it isn't a problem. Indeed, the fact that it's a > >hard problem is why it hasn't been solved yet, and why the best idea > >anyone has come up with is sheer persistence over time. > > Yupolutely! And in May of 1994 I came back from meeting Linus Torvalds > and made a presentation to my Digital Unix management at DEC that had as > a final bullet on the last page: > > o Linux is inevitable! > > They asked me what that bullet meant, and I said that no one could stop > Linux. My management laughed. > > Now Digital Unix is dead and most of them work for Red Hat. > > What I meant by "Linux is inevitable" was that the concept of designing > a FOSS ecosystem with community was inevitable. "Linux" itself may > migrate, fork, evolve, etc. but the model is here, and it will > accelerate. > > When I worked for Bell Labs in 1982 I heard someone say "I do not know > what the next operating system will be, but I bet it will be based on > Unix." I answered "I don't know what the next operating system will be, > but I bet it will be called Unix", meaning that the operating system > would evolve maintaining the same name. I was only wrong by a couple of > letters. :-) > > md > > ___ > gnhlug-discuss mailing list > gnhlug-discuss@mail.gnhlug.org > http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/ > __
Re: Interesting article,
O.K., I will wade in here. :-) For the most part, Ben is right. Vendors who completely control both hardware and software can make the "best" products, if your definition of "best" is a limited market of items, and you are willing to pay for them. MVS, VMS, Digital Unix. Rock solid, stable, scalable. REALLY EXPENSIVE. A lot of Microsoft's problems have to do with drivers that come from different vendors, trying to control different controllers that fit into buses that are not that well documented. One mistake in a driver (inside the monolithic kernel) and BAM! Lockup and blue screen. I am amazed that Microsoft's eco-system can actually boot at all. >> Apple could have >> crushed MS by now if they had gone with the GPL attitude instead of >> picking BSD so they could keep all their toys to themselves. >Yah, I'm not buying that. If all you needed was the GPL, Linux >would already have crushed Microsoft. Ben, I think you are under-estimating your own argument about inertia. Inertia is all about acceleration, not really speed. In 1991 Microsoft was already going 50,000 mph and accelerating and Linux started from zero, with almost zero acceleration. In 1994 a lot of the vendors seemed as if they were going to give the server market to WNT. There were a lot of people saying that "Unix was dead". After twenty years Microsoft is still accelerating, but I think it is accelerating at a slower pace, and FOSS is accelerating at a faster pace, but has still not caught up. Then there is distance traveled, or "speed over time" (in this case, installed base). It may take a very, very long time before FOSS has the same installed base, much less "crushing Microsoft". Apple has existed for about the same time as Microsoft, and still has about the same market penetration as twenty years ago. Its acceleration is a lot slower, and more or less allowed it to keep the same desktop and server market share (or maybe lost server market share in that time). >(When Linux first came out, you also had a fleet of commercial Unixes, >Novel, several BSDs, OS/2, BeOS, and all sorts of other bit player >platforms. Today it's all Microsoft, with Apple and Linux nipping at >their heels. Sadly, and from a "choice" and "research" viewpoint this is true. But the fragmentation meant that unless any of them reached critical stage, they would be just what you said "bit players", and would have died anyway. > Absolutely correct! However, the fact that's it's a hard problem to >solve doesn't mean it isn't a problem. Indeed, the fact that it's a >hard problem is why it hasn't been solved yet, and why the best idea >anyone has come up with is sheer persistence over time. Yupolutely! And in May of 1994 I came back from meeting Linus Torvalds and made a presentation to my Digital Unix management at DEC that had as a final bullet on the last page: o Linux is inevitable! They asked me what that bullet meant, and I said that no one could stop Linux. My management laughed. Now Digital Unix is dead and most of them work for Red Hat. What I meant by "Linux is inevitable" was that the concept of designing a FOSS ecosystem with community was inevitable. "Linux" itself may migrate, fork, evolve, etc. but the model is here, and it will accelerate. When I worked for Bell Labs in 1982 I heard someone say "I do not know what the next operating system will be, but I bet it will be based on Unix." I answered "I don't know what the next operating system will be, but I bet it will be called Unix", meaning that the operating system would evolve maintaining the same name. I was only wrong by a couple of letters. :-) md ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
Re: Linux for "cloud computing": Request for Input
On Fri, Mar 5, 2010 at 12:46 PM, Jon 'maddog' Hall wrote: > Any other ideas on the topic of "Why Linux for Cloud Computing?" Virtualization? > Any blatant negatives for Linux as a platform? Too many choices. Once you've chosen linux over other options, you've still got a ton of decisions to make. You mention choosing between security models as though this is a good thing. Maybe it's not [1]. Choosing a different platform might mean you have to make fewer decisions. (This isn't specific to cloud computing.) [1] For example, see http://www.neurosciencemarketing.com/blog/articles/more-choices-fewer-sales.htm -Brian ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
Re: Interesting article,
On Fri, Mar 5, 2010 at 1:22 PM, Benjamin Scott wrote: > On Fri, Mar 5, 2010 at 10:41 AM, Alan Johnson wrote: > >>> > Well, except MacOSX has specific hardware. > >>> > >>> Indeed, that's a big part of Apple's strategy. Design the hardware > >>> and the software together, and they'll work well together. And there > >>> is something to be said for that. > >> > >> And that's one of Apple's prime advantages. > > > > Advantage? Well, for Apple, it is an advantages over MS, but certainly > not > > for the users. > > Eh, I'm not sure about that. If you buy from Apple's extremely > limited pool of products, things tend to work together far better than > I've seen on any Microsoft-compatible platform, even if you buy from a > single vendor. > > This follows with Sun products too. Solaris on a Sun x86 workstation works very well & everything works. Put Solaris on a Dell and your mileage may vary with network, sound and graphics drivers. Linux has more drivers ported and will usually just work where Solaris won't. I have a Linux box at home to drive my SCSI tape drive because none of the older SCSI PCI adapters I have are supported in 64 bit Solaris. Some might be supported in 32 bit Solaris. Since they're older, Sun and the vendors don't have much interest in porting the drivers. There are a few open source drivers out there, but nowhere near what Linux has. ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
Re: Interesting article,
On Fri, Mar 5, 2010 at 10:41 AM, Alan Johnson wrote: >>> > Well, except MacOSX has specific hardware. >>> >>> Indeed, that's a big part of Apple's strategy. Design the hardware >>> and the software together, and they'll work well together. And there >>> is something to be said for that. >> >> And that's one of Apple's prime advantages. > > Advantage? Well, for Apple, it is an advantages over MS, but certainly not > for the users. Eh, I'm not sure about that. If you buy from Apple's extremely limited pool of products, things tend to work together far better than I've seen on any Microsoft-compatible platform, even if you buy from a single vendor. The IBM pee sea is a loose collection of vaguely similar things which happen to work together sometimes -- and that's being kind. Even if you buy everything from a single vendor, things rarely work as well together as they do when one company designs everything. The IBM-PC platform was not designed -- it evolved. Like the house that Jack built, things have been stuck on, later removed, changed, modified, extended, and reinterpreted so many times, by so many different actors, it's a wonder it stays standing. The single-vendor solution has to be built to work in that environment, and that's harder to do. In contrast, the Apple dictatorship does mean that standards are actually... well, *standard*. Look at Jerry Feldman's problems with partitioning. Ask three different programs how to do partitioning on an IBM-PC, and you'll get at least four different, mutually incompatible, data-destroying answers. In the People's Republic of Cupertino that would never happen. Whatever Apple decrees is The One True Way to do things. Does this limit freedom? Absolutely! Would I, personally, give up that freedom to get the uniformity of design? No way! But for someone who's interests consist entirely of surfing the web, listening to music, and doing the occasional resume or spreadsheet -- they see a real appeal in having someone else do all your thinking for them. As Larry Wall said about Perl, "Perl doesn't stop you from doing stupid things, because that would also stop you from doing clever things". Well, Apple stops their platform from doing stupid things. As with any dictatorship, things are great as long as you happen to agree with the dictator. But if one day the dictator decrees that computers should only be available with glossy screens, and you want a matte screen... well, too bad for you. Or: Apple no longer makes printers. So if you want to print from your oh-so-perfect Mac, you have to leave the Apple utopia, and come back to the ecosystem of the mass market. That's something Apple doesn't mention in their advertisements. Which is interesting, because their target market includes a lot of people who like to print everything... > The only reason Windows has dominance over Linux is inertia ... Just as the only reason mankind is limited to one planet is inertia. > ... and the only reason Apple has dominance over Linux is marketing. Certainly, Apple's marketing is brilliant. They know exactly what people want to hear, and they say it. But, in all fairness, they also steer their ship in that direction as well. They see a lot of people frustrated with the pee sea, and they build their platform -- the car with the hood welded shut -- specifically to appeal to that crowd. > Apple could have > crushed MS by now if they had gone with the GPL attitude instead of picking > BSD so they could keep all their toys to themselves. Yah, I'm not buying that. If all you needed was the GPL, Linux would already have crushed Microsoft. It's had 20 years and Microsoft has only gotten stronger in that time. (When Linux first came out, you also had a fleet of commercial Unixes, Novel, several BSDs, OS/2, BeOS, and all sorts of other bit player platforms. Today it's all Microsoft, with Apple and Linux nipping at their heels. If I were to draw a conclusion from that, it would be that GPL is good for crushing bit players, not the big guys.) > To be clear in limiting my own zealotry, I concede that there are a number a > specific use cases that are better addressed by Windows or Apple. However, > the vast majority of those are catch-22's like the ones described in the > article that started this thread. Absolutely correct! However, the fact that's it's a hard problem to solve doesn't mean it isn't a problem. Indeed, the fact that it's a hard problem is why it hasn't been solved yet, and why the best idea anyone has come up with is sheer persistence over time. -- Ben ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
Linux for "cloud computing": Request for Input
Hi, I am doing some consulting about "Why Linux is good for cloud computing" (and for that matter "Software as a Service" (SaaS), which I consider more or less one and the same). I am going to start with the fact that Unix systems were designed (almost) from the beginning to be multi-tasking and multi-user, then worked its way out to be multi-architecture, multi-threaded and embraced 64-bit systems relatively early. In addition, Unix systems utilized network processes and stressed client/server models (evident in daemons, NFS, the X Window System), as well as scripting languages that allowed control of applications through APIs and not just a graphical interface. This gave: o scalability o some architectural security o standards based development A lot of work in highly available servers was done with Unix systems such as Solaris, Digital Unix and others. Linux, patterned after Unix, inherited a lot of these characteristics. In addition with Linux you get: o "Open Source" that helps give: - needed bug fixes rapidly (under control of the service provider) - development of new features by large numbers of programmers, researchers, etc. - many "middleware" and emerging "management" systems are being developed on Linux, or developed as FOSS projects + Eucalyptus + Languages like Ruby, PHP, Python - licensing terms that do not restrict what you can offer to customers (i.e. how many instances can you run, how many customers can attach, etc.) - a couple of different security models (SE/Linux, AppArmor, as examples) to chose from - graceful degradation: If a technology is abandoned, the provider of services can maintain it until a migration can occur through community action - Open development model - allows service providers to plan ahead and have input to development o Linux also was used as the basis for Beowulf systems, which developed a lot of code surrounding "high performance clusters", leading to highly scalable systems o Basically can be same OS on desktop as servers Any other ideas on the topic of "Why Linux for Cloud Computing?" Any blatant negatives for Linux as a platform? Note that I am not arguing for or against "the cloud", just why Linux is or is not a good system for it. The entire paper will be available for free download when it is finished. Thanks in advance for your input. md ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
Re: Interesting article, games
On Fri, Mar 5, 2010 at 12:31 PM, Benjamin Scott wrote: > On Fri, Mar 5, 2010 at 11:18 AM, Thomas Charron wrote: >> Except for the fact that Andriod is pretty much all written in Java. >> :-D And doesn't use X. And must be run in a VM which isn't the Java >> VM. > Oh. I didn't know that. I don't have much interest in mobile phone > development; I just keep seeing *nix people going on and on (and on > and on and on) about Android. I foolishly ASSumed it had something to > do with *nix. When I jumped in and started writing apps, I was really disappointed I couldn't just compile and run them natively. It's all open, don't get me wrong. But there's a whole slew of custom-google-glue-and-tweaks, which was disappointing. -- -- Thomas ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
Re: Will Android draw developers to Linux? (was: Interesting article, games)
On Fri, Mar 5, 2010 at 12:05 PM, Joshua Judson Rosen wrote: > BUT: I wonder if maybe Ben isn't talking about it being a `gateway drug' > that draws people to platforms that are *technologically similar*, > but if he instead is talking about it drawing people to platforms > that are more, um..., `culturally' related? Yah! That's what I meant! Exactly! What he said! ;-) -- Ben ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
Re: Interesting article, games
On Fri, Mar 5, 2010 at 11:18 AM, Thomas Charron wrote: > Except for the fact that Andriod is pretty much all written in Java. > :-D And doesn't use X. And must be run in a VM which isn't the Java > VM. Oh. I didn't know that. I don't have much interest in mobile phone development; I just keep seeing *nix people going on and on (and on and on and on) about Android. I foolishly ASSumed it had something to do with *nix. -- Ben ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
Will Android draw developers to Linux? (was: Interesting article, games)
Thomas Charron writes: > On Fri, Mar 5, 2010 at 9:52 AM, Benjamin Scott wrote: > > > > Droid seems like the first Android phone that's getting > > marketing budget, mainstream attention and adoption, and also has an > > impressive feature set. While widespread use of Android as a generic > > phone engine might be good for Linux tech, it's not going to help the > > desktop market much because it's not going to be attracting > > application programmers. But if Droid gathers the kind of developer > > ecosystem that iPhone has, *that* might be a gateway drug for the > > Linux desktop. > > Except for the fact that Andriod is pretty much all written in Java. > :-D And doesn't use X. And must be run in a VM which isn't the Java > VM. So, from everything that I remember reading, it's basically *not Java*, but rather some weird mutant mostly-similar Java-alike. But at least it's closer to Java than the underlying OS is to anything that people call "Linux". Have you read the recent "Android Mythbusters" discourse? Even the kernel (Linux proper) is significantly mutant. BUT: I wonder if maybe Ben isn't talking about it being a `gateway drug' that draws people to platforms that are *technologically similar*, but if he instead is talking about it drawing people to platforms that are more, um..., `culturally' related? Almost like how people buy Macintoshes because they support their iPods better than Windows PCs would? -- "Don't be afraid to ask (λf.((λx.xx) (λr.f(rr." ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
Re: Interesting article, games
On Fri, Mar 5, 2010 at 9:52 AM, Benjamin Scott wrote: >> You mean Android right? Droid is one model of phone using the Android >> platform. > Yes and no, because... >> I've read of fragmentation in the Android market. > Exactly. Droid seems like the first Android phone that's getting > marketing budget, mainstream attention and adoption, and also has an > impressive feature set. While widespread use of Android as a generic > phone engine might be good for Linux tech, it's not going to help the > desktop market much because it's not going to be attracting > application programmers. But if Droid gathers the kind of developer > ecosystem that iPhone has, *that* might be a gateway drug for the > Linux desktop. Except for the fact that Andriod is pretty much all written in Java. :-D And doesn't use X. And must be run in a VM which isn't the Java VM. -- -- Thomas ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
Re: Interesting article,
On Fri, Mar 5, 2010 at 8:52 AM, Tom Buskey wrote: > > Well, except MacOSX has specific hardware. >> >> Indeed, that's a big part of Apple's strategy. Design the hardware >> and the software together, and they'll work well together. And there >> is something to be said for that. >> > > And that's one of Apple's prime advantages. > Advantage? Well, for Apple, it is an advantages over MS, but certainly not for the users. This goes back to my post about Italian Trains. Apple has managed to convince the bulk of their user base that legally limiting the hardware they can use is some how to the advantage of the user, when really it is only to the advantage of Apple in that it makes their job easier and they can charge more based on perceived value. The user can get the same "advantages" of hardware limitation from Linux or Windows by only buying from a well run OEM that knows what hardware works best. For the slightly more advanced user, they can limit their hardware choices to those that have been vetted by the larger community, as I and many on this list do. Because of dominance, Windows does not have to work as hard at this, but both Windows and Apple at least TRY to allow more freedom in hardware for their users. The only reason Windows has dominance over Linux is inertia, and the only reason Apple has dominance over Linux is marketing. Apple could have crushed MS by now if they had gone with the GPL attitude instead of picking BSD so they could keep all their toys to themselves. They are stomping on the shoulders of giants rather than standing. To be clear in limiting my own zealotry, I concede that there are a number a specific use cases that are better addressed by Windows or Apple. However, the vast majority of those are catch-22's like the ones described in the article that started this thread. E.g. poor gaming and OEM adoption implies low market share which implies poor gaming and OEM support. Both issues are being whittled away at and also remember that the global market share of Linux is much greater than in this country and there should be no confusion about the correlation to the location of the MS and Apple headquarters. China and India have no national interest in choosing MS or Apple, and they both have a whole lot more potential users. This is why Google is getting in to the hardware market with Linux (Android). They don't want to sell a few hundred thousand devices in the US. That's just a testing ground. They want to sell billions of devices globally. MS and Apple will never achieve that with closed technologies because they could never handle the overhead of maintaining translations, etc. The powerful concept is that Google does not even have to make any money on the hardware, as long as they don't loose much, and users keep using "google" as a verb. =) That said, I am completely convinced that the sum of the *nix flavors out there are surely superior in an over-all sense when compared to the sum of Windows and Apple options. (I know, I know; MacOS is a *nix, but let's split out for reasons mentioned above, shall we?) The only hold up at this point is strong enough business, or small set of businesses, getting behind it to ease the average user into it. When Dell and/or HP start offering Linux on the majority of their hardware, that will be the beginning of the end of closed systems like Windows and Mac. When Google puts a billion Android devices in the global market, I won't have any trouble playing downloaded Amazon video files on my Ubuntu box hooked up to my TV. For now, I'll have to settle for Redbox over Netflix (since their streaming player requires Windows even though their set-top box runs Linux) and playing videos I buy from Amazon through their streaming player on my TV box while it also functions as commercial grade backup server, carrier grade firewall/NAT-router/traffic shaper, local/Internet streaming media server/player, educational tool for my children, smart home server/interface (if i ever get around to it), and yes, with plenty of selection for my limited tastes as a recovering gamaholic, a gaming machine. Oh, woe is me. ;-) ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
Re: Interesting article, games
On Fri, Mar 5, 2010 at 9:23 AM, Tom Buskey wrote: >> I remember that! Computers with the hottest graphics hardware on >> the planet, and Doom still did all rendering in software and just >> blitted bitmaps to X. :) > > Didn't Doom use OpenGL as its engine? Id is one of the reasons DirectX > didn't wipe out OpenGL. Nope, Doom used a one-off, all-software rendering engine. Doom ran on MS-DOS, in the days of Windows 3.x, before 3D hardware acceleration was really available on the IBM-PC. Heck, the fact that it used i386 protected mode was something of a big leap forward. I still remember the "DOS/4GW" banner from startup. One of the reasons Doom received so much praise is how it managed to get so much done on a fairly craptacular OS and hardware platform. I seem to recall once reading that the id Software level designer ran on SGI's using OpenGL, and that was initially a barrier to bringing a designer to the consumer market. Third-parties ended up building stripped down map editors -- they just drew a top-down line map instead of visualizing, but they could run on the pee sea. >> > That mainstream OS is probably the iPhone OS, not MacOSX. >> >> I wonder, if Droid takes off like iPhone did, will that bring more >> interest to Linux? > > You mean Android right? Droid is one model of phone using the Android > platform. Yes and no, because... > I've read of fragmentation in the Android market. Exactly. Droid seems like the first Android phone that's getting marketing budget, mainstream attention and adoption, and also has an impressive feature set. While widespread use of Android as a generic phone engine might be good for Linux tech, it's not going to help the desktop market much because it's not going to be attracting application programmers. But if Droid gathers the kind of developer ecosystem that iPhone has, *that* might be a gateway drug for the Linux desktop. -- Ben ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
Re: Interesting article, games
On Fri, Mar 5, 2010 at 9:14 AM, Benjamin Scott wrote: > On Fri, Mar 5, 2010 at 9:02 AM, Tom Buskey wrote: > > I've played Doom on SGI Irix FWIW. Doom was everywhere. > > I remember that! Computers with the hottest graphics hardware on > the planet, and Doom still did all rendering in software and just > blitted bitmaps to X. :) > Didn't Doom use OpenGL as its engine? Id is one of the reasons DirectX didn't wipe out OpenGL. > > > That mainstream OS is probably the iPhone OS, not MacOSX. > > I wonder, if Droid takes off like iPhone did, will that bring more > interest to Linux? > > You mean Android right? Droid is one model of phone using the Android platform. I've read of fragmentation in the Android market. If developers can't depend on a consistant sound/graphics/accelerometer, they're not going to develop for it. How many games are on WINCE/Windows Mobile which has a similar problem? ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
Re: Interesting article, games
On Fri, Mar 5, 2010 at 9:02 AM, Tom Buskey wrote: > I've played Doom on SGI Irix FWIW. Doom was everywhere. I remember that! Computers with the hottest graphics hardware on the planet, and Doom still did all rendering in software and just blitted bitmaps to X. :) > That mainstream OS is probably the iPhone OS, not MacOSX. I wonder, if Droid takes off like iPhone did, will that bring more interest to Linux? -- Ben ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
Re: Interesting article, games
On Thu, Mar 4, 2010 at 10:48 PM, Ben Eisenbraun wrote: > On Thu, Mar 04, 2010 at 06:12:53PM -0500, Star wrote: > > There have been a couple of great releases specifically targeting Linux > > as a platform. I'm thinking of Unreal, EVE, and Farcry (i think?) > > There have been some. I remember playing Tribes 2 during lunch on the > linux workstations in the NOC. Id has historically been good about > releasing its games for Linux, but then the driver problem arises. Even > with the binary Nvidia driver, you can expect a 25-50% drop in frame rate > on the same hardware switching between linux and Windows. > I've played Doom on SGI Irix FWIW. Doom was everywhere. > > There's just not enough of a market for linux games for the publishers to > port their games and for the hardware manufacturers to spend time tuning > their drivers for linux. > > Don't most publishers focus on the consoles? Playstation, Xbox, Wii, DS, PSP and iPhone? The end platform is consistant and not changing. Anyway, I think if Apple makes it as a gaming platform, it will be because > they made it as a mainstream OS first. (Or possibly because they used the > mobile gaming arena as the fulcrum upon which they levered game developers > on to OS X.) > > That mainstream OS is probably the iPhone OS, not MacOSX. ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
Re: Interesting article,
On Thu, Mar 4, 2010 at 6:00 PM, Benjamin Scott wrote: > On Thu, Mar 4, 2010 at 2:37 PM, Tom Buskey wrote: > >> Freedom, and low cost, and robustness, and > >> security, and choice, and all that good stuff. > > > > s/Linux/MacOSX/ > > Yes and no. MacOS X certainly has a number of selling points, > several of which it shares with Linux, others which are unique. Apple > has done a very nice job with human factors engineering, OS features, > their bundled applications. Security is pretty good and certainly the > OS itself is robust. > > But low cost? Freedom? You never really own a Mac -- you're just > renting it from Steve Jobs. As someone said to me recently, "There > can be more than one evil empire." > > I don't disagree :-) My point was with zealotry and converting people from windows to another platform. The layman will have similar objections in switching to Linux or MacOSX. Linux has the Freedom argument in addition. Apple is more of a monopoly then Microsoft IMO. > > Well, except MacOSX has specific hardware. > > Indeed, that's a big part of Apple's strategy. Design the hardware > and the software together, and they'll work well together. And there > is something to be said for that. > And that's one of Apple's prime advantages. ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
[OT] Re: Interesting article, games and ugly pants
>ugly pants carrying Windows laptops around. I must admit I never related ugly pants with Windows laptops. I sense a follow-up study, but firstwhat is the definition of "ugly pants"? md ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
