Linux-Advocacy Digest #443, Volume #30 Sun, 26 Nov 00 14:13:04 EST
Contents:
Re: Windoze 2000 - just as shitty as ever (T. Max Devlin)
Re: Windoze 2000 - just as shitty as ever (T. Max Devlin)
Re: C++ is very alive! (mlw)
Re: Linux growth rate explosion! (mark)
Re: Of course, there is a down side... (mark)
Re: Response to: MS Office sucks? So why is anyone using it? (mark)
Re: Response to: MS Office sucks? So why is anyone using it? (mark)
Re: Mandrake 7.2 Quick Review (.)
----------------------------------------------------------------------------
From: T. Max Devlin <[EMAIL PROTECTED]>
Crossposted-To: alt.destroy.microsoft,comp.os.ms-windows.advocacy
Subject: Re: Windoze 2000 - just as shitty as ever
Date: Sun, 26 Nov 2000 12:40:48 -0500
Reply-To: [EMAIL PROTECTED]
Said mark in alt.destroy.microsoft on Sat, 25 Nov 2000 22:02:05 +0000;
>In article <8vpb92$5c60s$[EMAIL PROTECTED]>, Ayende Rahien wrote:
>>
>>"Curtis" <[EMAIL PROTECTED]> wrote in message
>>news:[EMAIL PROTECTED]...
>>> mark wrote...
>>
>>
>>> > Linux at some undetermined point after 1996 and
>>> > up to and including today, (wonder which version & which
>>> > distro?), and now Win2k with its somewhat restricted set of
>>> > available apps?
>>>
>>> Restricted apps for Win2k. Hehehehe. What apps are you looking for that
>>> gives you this impression? Or is this second hand information that you're
>>> stating?
>>
>>There are all too many application that are written by lazy/idiot
>>programmers which assume 95/98/ME and full access to the registery.
>
>The thread was about a claim of some applications which were
>available in 1996 for win9x, had some functional equivalent at
>the same time for OS/2, had some functional equivalent at
>some indetermined later date on Linux and also have some
>functional equivalent now on Win2k (or they've been ported).
No, the thread was about many things. This point in particular was a
straw man, I have to admit, that you brought up in discussion with
Curtis when he described his experience with PC OSes.
--
T. Max Devlin
*** The best way to convince another is
to state your case moderately and
accurately. - Benjamin Franklin ***
Sign the petition and keep Deja's archive alive!
http://www2.PetitionOnline.com/dejanews/petition.html
====== Posted via Newsfeeds.Com, Uncensored Usenet News ======
http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
======= Over 80,000 Newsgroups = 16 Different Servers! ======
------------------------------
From: T. Max Devlin <[EMAIL PROTECTED]>
Crossposted-To: alt.destroy.microsoft,comp.os.ms-windows.advocacy
Subject: Re: Windoze 2000 - just as shitty as ever
Date: Sun, 26 Nov 2000 12:50:42 -0500
Reply-To: [EMAIL PROTECTED]
Said Ayende Rahien in alt.destroy.microsoft on Sat, 25 Nov 2000 20:52:54
>"mark" <[EMAIL PROTECTED]> wrote in message
>news:[EMAIL PROTECTED]...
>> In article <8vmhss$50m41$[EMAIL PROTECTED]>, Ayende Rahien wrote:
>> >
>> >"Sigvaldi Eggertson" <[EMAIL PROTECTED]> wrote in message
>> >news:8vmffj$crq$[EMAIL PROTECTED]...
>> >> In article <8vlgh0$4tt57$[EMAIL PROTECTED]>,
>> >> "Ayende Rahien" <[EMAIL PROTECTED]> wrote:
>> >> >
>> >> > Most people in the world *don't* know english, therefor, they need an
>> >> OS in
>> >> > their own language, and they'll pay for it.
>> >> >
>> >> > Take a look at what happened when Iceland wanted windows in their own
>> >> > language, btw.
>> >> >
>> >> We,here in Iceland, know English (at least some) but what happened was
>> >> that we got Windows 98 in Icelandic a couple of years back.
>> >
>> >IIRC, there was a problem with that (it may be with win95, I'm not sure)
>> >that MS didn't want to localize windows because large part of the
>population
>> >had a good control in english.
>>
>> The actions of a monopoly
>
>No, the actions of a company whose interest is in profit.
But I would suspect that, since this was reported on, it involved many
more users than some of the other languages that Windows supports. So
is it profit (the cost of production versus the potential market) or is
it monopolization (the greatest number of users locked in) which
determined whether Icelandic, or something like, say, Bali, was
supported?
>It wasn't worth it to localize Windows, because most of the people could use
>the english version.
What difference at all does that make? Does a producer base its
decision on what to invest in on whether it will increase their sales,
or whether it will "cover" a particular victim/consumer? Wouldn't
whether Icelanders would buy a non-english version be the question,
rather than whether they can use an english version? Is MS trying to
increase their sales, or are they trying to monopolize the market? You
tell me.
--
T. Max Devlin
*** The best way to convince another is
to state your case moderately and
accurately. - Benjamin Franklin ***
Sign the petition and keep Deja's archive alive!
http://www2.PetitionOnline.com/dejanews/petition.html
====== Posted via Newsfeeds.Com, Uncensored Usenet News ======
http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
======= Over 80,000 Newsgroups = 16 Different Servers! ======
------------------------------
From: mlw <[EMAIL PROTECTED]>
Subject: Re: C++ is very alive!
Date: Sun, 26 Nov 2000 12:55:20 -0500
Russ Lyttle wrote:
>
> mlw wrote:
> >
> > Charlie Ebert wrote:
> > >
> > > Aren't you people forgetting that HURD is based
> > > on C++ with that microkernel technology.
> > >
> > > They are going the be the C++ club of the future.
> > >
> > > C++ is far from dead. It's just not being used
> > > as few old timers really understand how to use it.
> >
> > !!! Warning!!! The following reads like flame bait, to many it may be.
> > It is not intended to be. !!
> >
> > I know lots of engineers that use C++.
> >
> > If I should be so bold, software engineers use tools like C++,
> > programmers use stuff like Java or VB. There is a difference, and it is
> > important to remember.
> >
> > What is an AVL tree, can you write one?
> Adelson-Velskii & Landis tree. Yes
> > What is a hash table, how would you write one?
> > What is a radix?
> Also known as "base", as in base-10, base-16, base-2, etc.
Actually, I was thinking more radix tree and searches. But, you are
right.
> > How does bsearch work? Could you write one?
> Binary Search? Binomonial-search? Yes
> > What is a histogram? Can you write one?
> A graph to represent data after it has been sorted into bins. Because a
> histogram is a graphical representation, it doesn't seem quiet right to
> say you would "write" a histogram. But we are probably talking semantic
> opinions.
Write a histogram system to perform analysis on a stream of data.
> > When is qsort a bad choice?
> Most of the time. But it is really bad if the partitioning routine
> divides the data into sets of 1 element and n-1 elements.
I almost never use qsort.
> > What other types of sorting algorithms are there, and what are the
> > pros/cons?
> Lots. merge sort, insertion sort, radix sort, etc. Each has
> characteristics that depend on the data.
> > What would you a histogram for?
> Back to the semantics? Any way, putting data into bins is often used in
> anything involving statistical analysis : Kalman filters, image feature
> detection, etc.
>
> > What would you use an AVL tree for? How about a hash table?
> When some insertion cost can be traded for good lookup speed. Hashtables
> can be used to save/lookup data if collisions are rare (but you trade
> memory for speed on sparse tables).
> > How does a linked list work? How about a double linked list?
> Node has pointer to child node. Node has pointers to child and parent
> node.
or left-right, prev-next, but yes.
> > How do you simulate subtraction with addition?
> twos-complement and add.
> > Which can be written more efficiently, divide by 3 or divide by 4?
> Normally divide by 4. But processors with a built in DIV operator might
> not show any difference.
> > On a pentium which is more efficient integer arithmetic or or floating
> > precision? How about PIII? Why?
> Don't have a book on the Pentium or PIII. On most processors of that
> general architecture, integer arithmetic is faster because the floating
> point divide doesn't work well.
> TMS320 series, for example, can do floating point multiply and add in
> one clock cycle, but ends up using over 100 to do a divide.
Actually, in classic processors, integer arithmatic is almost always
faster. However, the latest versions of processors are so closely linked
to the FPU, and the FPUs are so fast, that floating point is almost
always faster now. Look at the MMX class of instructuions, it is using
the FPU to do stuff.
> > What is the advantage of keeping objects in memory closer to one
> > another? When/how is this a problem in a multitasking multiprocessor
> > environment?
> Depends on objects. In multitasking-multiprocessing systems, objects
> need to be confined to the memory block they are associated with. You
> never want one process trying to directly access objects in another
> process memory space.
Actually is is a two way problem. You want to keep objects close
together to keep it all in motherboard cache, but you need to make sure
that on a multiprocessor system that you do not force cache coherency
issues at trying to access data forced in each CPU's cache, causing
"false parallelism."
> > What are the trade-offs between fixed memory block allocation vs
> > variable block?
> You might not find memory for variable block allocation, but fixed block
> might allocate memory that is never used.
> > How about first fit, last fit, or best bit allocation strategies?
These are pretty common terms in the memory management world:
First fit:
First free block that satisfies an allocation request.
Last fit:
Last free block that satisfies an allocation request.
Best fit:
The smallest block the satisfies an allocation request.
Don't bother, each one can be argued as better than the other. There are
no right answers. It depends largely on how your application behaves.
Whole chapters of operating system design books are filled with
dicussions about this topics.
> > These are all very important questions (and only scratching the
> > surface). If you do not know these sorts of things cold, then you are a
> > programmer and are probably better off doing stuff in Java or some other
> > limited environment until you learn more. It isn't until you understand
> > these sorts of things do you understand how to develop software.
> >
> Except for the last item and the Pentium specific question, I work with
> those items every day. Implementing them has nothing to do with choice
> of language. You can do all of them in any complete language : assembly,
> Ada, C, C++, Java, Pascal, COBOL, FORTRAN, BASIC, whatever. In fact I
> have written them in all of the above except COBOL. The advantage of
> Ada, C, and Java is that once written in those languages, I can use them
> almost anywhere. Java is the best for networks, but Ada and C are best
> for microcontroller/DSP targets. C++ is best if you have a good process,
> want to hire cheap programmers(Ada programmers are expensive!), and lots
> of memory is available .
Lets not get into the C++ thing again, we have to agree to disagree
about that. Geez, if you came in for an interview, you'd have to be a
psychopath not to get hired. Suffice to say I am impressed. Most guys
don't get it.
>
> > I am aware that many will be pissed off by this attitude, but in
> > defense, it should be noted that a good number of people take software
> > development seriously. We spend years learning and improving our skills.
> > It takes an amount of dedication to get to the point where the question
> > about how to do something is a no brainer, and the tricks are purely
> > implementation issues on ever changing platforms.
> >
> > C++ isn't dead, far from it. It is a power tool. people with few skills
> > are scared of powertools. Craftsmen love powertools because they know
> > how to use them, they make the job easier and, in the end, make a better
> > product.
> >
> People with real understanding use the best tool for the job. Define
> your requirements first. Then do a preliminary design. Next, do a trade
> study and pick your tools. After all, a "Power" saw isn't very good at
> pulling a plow on a hillside.
No, but a good tractor is, so why waste time with an ox?
>
> > --
> > http://www.mohawksoft.com
>
> --
> Russ Lyttle, PE
> <http://www.flash.net/~lyttlec>
> Not Powered by ActiveX
--
http://www.mohawksoft.com
------------------------------
From: [EMAIL PROTECTED] (mark)
Crossposted-To:
comp.lang.java.advocacy,comp.os.ms-windows.advocacy,comp.os.ms-windows.nt.advocacy
Subject: Re: Linux growth rate explosion!
Date: Sun, 26 Nov 2000 18:17:30 +0000
In article <8vr8r9$5a7fd$[EMAIL PROTECTED]>, Ayende Rahien wrote:
>
>"mark" <[EMAIL PROTECTED]> wrote in message
>news:[EMAIL PROTECTED]...
>> In article <8vqvqq$5f036$[EMAIL PROTECTED]>, Ayende Rahien wrote:
>> >
>> >"mark" <[EMAIL PROTECTED]> wrote in message
>> >news:[EMAIL PROTECTED]...
>> >> In article <qFZT5.10131$[EMAIL PROTECTED]>,
>> >> Chad Mulligan wrote:
>> >
>> >
>> >> >Because they are still looking for a way in.....
>> >>
>> >> Yeah, on windows trojan ports - as I said above. I don't run
>> >> windows so the trojans they're looking for are not on my
>> >> machine.
>> >
>> >The most wide spread OS is?
>> >Most trojans are written to what OS?
>> >
>> >Combine those two answer and you'll realize why you've a lot of port
>scans
>> >to windows trojans
>> >Not to mention that this is about the most inefficent way to do this.
>>
>> Doh. How do you get a trojan onto a unix machine?
>
>Same mecanism you get one into a win machine.
>Lure the user to open it.
No, the user needs to save it, give it executable permissions,
su to root, give it root/suid permissions, put it into the path,
add a script into /etc/init.d or /etc/rc.d to get the trojan
started, modify the firewall scripts to open the required ports,
etc. etc. etc.
Note that 'open' just means 'open' in unix, it does not mean
run.
>
>
>> >> >> I see thousands of scans for windows machines. I don't see any for
>> >> >> unix machines.
>> >> >>
>> >> >
>> >> >Because they've already found a way in.
>> >>
>> >> No, not on my machine. I've never been cracked.
>> >
>> >That you know of :)
>>
>> mmmm :)
>> >
>> >
>> >> The massive amount of port-scanning shows how many thousands of
>> >> people are trying, continuously, to crack windows machines using
>> >> one of the very very many trojans out there. These are the ones
>> >> which are eg., regularly posted to newsgroups, knowing that many
>> >> windows users run Outlook express, and will 'open' the attachment,
>> >> infect their machine because 'open' also happens to mean 'run'
>> >> in the windows world.
>> >
>> >You can't tell the difference between a trojan and an exploit?
>> >If there was a wide spread trojan for unix machines, wouldn't people scan
>> >for it?
>>
>> How do you get a trojan onto a unix machine? You need a delivery
>> mechanism. Microsoft provides one. You need root access, Microsoft
>> gives that to everyone.
>
>
>What delivery mecanism MS provides?
>No, win9x gives that to everyone. And it's shit.
>Use NT or better yet, 2000 for a good system.
I don't get a choice. Myself and all my colleagues get
given Win98SE or previously Win95. The company employs
around 150,000 people.
Personally, I disabled visual basic ages ago (at least
the msoffice bit).
>
>
>> >If unix was a wide spread OS, used by average computer illeterate
>persons,
>> >you would see the same for unix.
>>
>> No, for the reasons I give above.
>
>Yes, for the reasons I give above.
No, it really isn't the same, is it. Microsoft provides the
delivery mechanism and the root access and the lack of firewalling
capability, along with a really obscure startup system (the
registry) which _anything_ can modify to startup itself at
startup.
Daemons (or whatever you call them in Win32 land) can even
prevent themselves from appearing in the task list.
>
>> >> Then, a few mins later, the scanner detects the running trojan,
>> >> and, guess what, your windows machine just got owned.
>> >
>> >That is true for all OS.
>> >
>>
>> No, for the reasons I give above.
>
>Yes, for the reasons I give above.
No for the reasons I give above. You really should learn
something about unix. You'll probably find yourself switching
to a decent OS when you realise just how had you've been.
>
>
Mark
------------------------------
From: [EMAIL PROTECTED] (mark)
Crossposted-To: comp.os.ms-windows.advocacy,comp.os.ms-windows.nt.advocacy
Subject: Re: Of course, there is a down side...
Date: Sun, 26 Nov 2000 18:22:59 +0000
In article <9G8U5.91$[EMAIL PROTECTED]>, PLZI wrote:
>
>"mark" <[EMAIL PROTECTED]> wrote in message
>news:[EMAIL PROTECTED]...
>> In article <V2_T5.242$[EMAIL PROTECTED]>, PLZI wrote:
>> >
>> >
>> >Want to reconsider "not useful at all"? Or care to explain how RPC or NFS
>is
>> >now suddenly not useful, nor open? Wake up. I'm not going to tell you how
>MS
>> >is suddenly all open and nicest guy on the block and the saviour of
>internet
>> >community, I'm just trying to explain that for once they did something
>which
>> >is open (to a degree, as RPC and NFS are "open") and useful.
>>
>> The items you've listed above were, as I recall, published _before_
>> they had been reverse engineered. I believe that they also are
>> sufficiently complete in themselves to be usable, whereas my
>> understanding is that the CIFS is not enough, on its own, to
>> implement the Microsoft file sharing and directory system, which as
>> far as I am aware has not been fully published, in 'infomrational'
>> form or otherwise. Unless you can find a reference to the rest of
>> if, of course?
>
>Now this is where the things get interesting. Now the samba people say, that
>they need to implement DC functionality (haven't really looked into this, but
>I presume they'd be interested in authentication/authorization mechanisms).
>These are not part of CIFS, they are part of networking functionality
>provided as a service, which happens to use CIFS (and other protocols).
>
>In essence, they are saying "Ok. GSM standard provides the protocol to move
>human speech. And now we have it. To make a phone, we still need to know how
>the phone works, how the compression algorithms and SIM card works. So now
>provide us the schematics of phone, source code to it, and the schematics of
>SIM card. Fail to do so, feel our wrath."
No, you totally misunderstand how telecoms standards work. It is
absolutely possible to design and build a functioning telecoms
network by doing everything (or the subset you want) of things
given in the ITU-T and ITU-R standards. ETSI standards equally
provide a useful overlay, including things like the fully open
GSM standard. Signalling is covered in Q.931 and related.
Transmission is covered in the G.7xx series, transmission performance
and synchronisation in the G.8xx series, maintenance in the M.xxxx
series, and so on. FYI I was rapporteur for Q.9/4 doing the
M.21xx and M.22xx series.
The standards document interfaces at each layer in order that every
phone can interwork via the air-interface with every other phone.
It has to be that way otherwise the bl**dy phone network wouldn't
work.
Unfortunately, there's a body of opinion in the computing
environment that believes this interoperability should be
optional in computing.
>
>I am NOT saying that MS is Doing the Right Thing when not giving the
>intricasies of these services - but I fully understand why they do not give
>them. I can come up with tens of examples, where the protocol is open, but
>the service which uses the protocol is prorietary.
>
>- PLZI
>
>
>
------------------------------
From: [EMAIL PROTECTED] (mark)
Subject: Re: Response to: MS Office sucks? So why is anyone using it?
Date: Sun, 26 Nov 2000 18:24:20 +0000
In article <8vqu6n$1q3$[EMAIL PROTECTED]>, Adam Warner wrote:
>Hi Mark,
>
>> I would recommend LDAP. It's free, supports multiple platforms,
>> debian has an implementation available. Why waste your country's
>> resources filling Bill's pockets when you could be educating your
>> upcoming youngsters with it as well?
>
>Does anyone know how hard/costly it would be to remove an ADS once
>implemented? (Maybe that's an unknown at this stage).
Trial systems and networks are cheap to remove.
>
>> >This is obviously not the only University/College/Institute where this
>will
>> >be occurring.
>>
>> Why? Personally, I'd be looking at options which leave more resources
>> for educating my people coming in.
>
>We agree about putting scarce University resources to the best possible use.
>I was only making an observation that Microsoft has probably been working
>with some other institutions around the world in a similar fashion.
I'm sure they're trying to corner as many as they can before
anyone calculates the cost.
Mark
------------------------------
From: [EMAIL PROTECTED] (mark)
Subject: Re: Response to: MS Office sucks? So why is anyone using it?
Date: Sun, 26 Nov 2000 18:25:34 +0000
In article <[EMAIL PROTECTED]>, Edward Rosten wrote:
>
>> I think that it won't happen without Microsoft effectively giving
>> the product to the universities, since what's available for
>> free is so good (don't forget that such as solaris are available
>> for free, as well as Linux and the BSDs and Minix). Apart from
>> a few exceptions (such as Oxford or Cambridge), most Universities
>
>There are some colleges which seem to be falling under their weight in
>gold. Not all colleges and departments are quite so lucky.
Sorry - I was a bit broad there.
>
>> are not exactly falling under the weight of their gold. A free
>> OS solution _must_ be worth considering for a non-profit body.
>
>
>There's plenty of free software flaoting around in Oxford. The
>engineering dept. has mainly PSs running Solaris (+ a few suns)
>avaliable for the student use. The physics department (IIRC) has ust got
>a load of new Linux machines. The big new mail server is a small Linux
>cluster, too. That said, there are site licenses for various bits of
>commercial software.
>
>
>
>-Ed
>
>
>--
>Did you know that the reason that windows steam up in cold | Edward
>Rosten
>weather is because of all the fish in the atmosphere? | u98ejr
> - The Hackenthorpe Book of lies | @
> | eng.ox.ac.uk
------------------------------
From: [EMAIL PROTECTED] (.)
Subject: Re: Mandrake 7.2 Quick Review
Date: 26 Nov 2000 18:40:27 GMT
[EMAIL PROTECTED] wrote:
> That's what I did but there is something wrong with the image.
No, there is something wrong with your brain. All of the rest of us
can make it work in a few seconds, claire. You cannot.
=====.
------------------------------
** FOR YOUR REFERENCE **
The service address, to which questions about the list itself and requests
to be added to or deleted from it should be directed, is:
Internet: [EMAIL PROTECTED]
You can send mail to the entire list (and comp.os.linux.advocacy) via:
Internet: [EMAIL PROTECTED]
Linux may be obtained via one of these FTP sites:
ftp.funet.fi pub/Linux
tsx-11.mit.edu pub/linux
sunsite.unc.edu pub/Linux
End of Linux-Advocacy Digest
******************************
