Mukund I think you're making several big mistakes: one of which appears to be in assuming that I'm being 'confrontational'.
> You are mistaken if you assume that anyone is *asking* you to adopt > Free/OSS software. We are not _asking_ you to use our software. You can > use whatever you want. There are several choices, even in the open > source realm (there's FreeNAS which was reviewed recently on Slashdot), > and of course there are the "big-name" commercial vendors such as NetApp > and EMC. Go ahead, use whatever you want. I think you mistake where I am, why I wrote what I did, and [perhaps not surprisingly] what my experience both of OSS and CSS software are. My comments were made entirely within the context of the question Rafiu asked, and should be read as such. As it happens, I've bench tested and/or deployed an awful lot of those commercial and "free" products myself, including both FreeNAS and OpenNAS; my comments were made in the light of that experience. > Free and open source software also come with no warranty, or guarantees > that they're fit for a purpose. Actually even the majority of commercial > software come with no warranty, but the free software camp is more vocal > in putting this up front. Thanks, as a developer [and vendor] of some 35 years standing with several CSS and OSS projects under my belt [and at various points several million seats] I think I just learned something: never make assumptions about who you're talking to. > The point I'm trying to make is that the choice must be yours and you > alone are responsible for any issues that crop up due to your use of > this software such as loss of data, loss of people's lives, etc. The risk management must certainly rest with the user; that doesn't mean that developers of OSS don't have obligations. Some are simply obligations to themselves to do the best they can. Others are much less well expressed and incohate. That responsibility on the part of developers towards "the public" is often not well understood or well expressed within or from the OSS community - it's still an emergent property of OSS and will almost certainly remain so for several years to come. Perhaps the /best/ expression of those responsibilities that /I/ know of can be found in Eben Moglen's DotCommunist Manifesto. If you don't know it then I strongly recommend that you take the trouble to become familiar with it. > What a support contract provides you from any vendor including Red Hat, > IBM, Novell, etc. is timely help for problems that are fixable by them. > If the problems are beyond their control, then they'll do everything > they can to get it resolved by those who can fix them, but that's where > the help ends. I can only repeat what I said before: it was /an/ /example/ of the ways in which the smaller businesses approach "purchases" of tools such as OF. On the one hand it's all about accessability, open-handedness, and responsiveness on the part of the development team, and on the other hand it's about encouraging potential new users to share the voyage with the community of developers and other users. Look very carefully at why Mozilla/Firefox/Thunderbird work for the general public, and why Kapor's OSSF Chandler project is still stuck in the mud after several years and much "expensive" thinking and development. > This "No-one got fired for buying IBM" statement is a relic of the past. > I guess you have never come across horror stories of how much money is > spent on TCO on an IBM service. I am not knocking them, but I'm saying > you never get more than what you pay for. You always get less, and > sometimes you get far less. And yes people do get fired for buying IBM, > and "nobody got fired for buying Openfiler (yet)". You've missed the point! Again. > > OF doesn't [yet] have that presence and reputation in the wider world - > > which makes the risks perceived by potential customers even higher. No > > matter what levels of support we as vendors and the OF dev team might > > provide, it's that customer inertia that has to be overcome. > > > > The risks lie upon the customer. The customer has to take decisions. > Openfiler comes with no warranty. Openfiler has been available as a > public release for over 2 years now, and as an internal company product > for over 3 years now. We ask all our customers (and yes we already have > "big-name" heavy-use customers too) to first evaluate the product > thoroughly in their networks, and only then decide if they want to use > it. Again, you mistake the point: it's not about actual risks, it's about /perceived/ risks. That's risks as seen by the potential users, and how they want ease of choice, familiarity, and often also simplicity. OF has none of those things - in particular it has no significant public market presence and no street credibility, unlike Firefox, Apache, MySQL etc. I am NOT suggesting that OF should not have that credibility; I'm suggesting the team needs to find ways to build it. >From that perspective those 'customers' have almost certainly never heard of OF but, depending upon where they sit in the marketplace, they will have heard of people such as Buffalo, Linksys, Netgear, Maxtor, Freecom, La Cie, and others who produce relatively inexpensive plug and go NAS devices. Such customers will "feel safe" with such choices because of the known name of the vendor and their continual presence in ads in magazines and on the shelves of high-street stores etc. Or. if they're at the upper end of the market, they will have heard of the products from all the major suppliers including e-NAS, EMC/Clarion and even the products from HP, IBM and MS - these customers will also "feel safe" making such decisions regardless of how valid their judgement is. All these customers' judgements aren't judgements of the qualities of the product, they're judgements of the company and how it portrays itself. I repeat this for you. The warranties aren't the issue here; it's about the publicly perceived qualities of the product and the product team. The street credibility of the product and the people behind that product. Perl and Apache and Berkeley and Zope and many other OSS "products" are perceived as safe bets /because/ of the qualities of the people behind them and the manner in which they present themselves. OF hasn't /yet/ reached that mass-market standing and it very very much *needs* to do so. The product is good enough to support that standing! [That's a positive compliment - take it as applause!]. However from where I sit the *business* practice isn't as good as the product. That's what I've been commenting on: the business development. > How is it premature? We are actively getting calls from vendors to have > a services section and to offer support, as their customers ask for > support and the vendor doesn't know anything about the product to > support it. When there are such requests, then the time _is_ right to > have support levels. Should we turn a deaf ear to these requests? Not at all; go ahead and provide such support. I'm talking about developing the 'business' aspects further - about finding more ways to engage a more diverse group of users. That in turn requires the product itself to be more engaging - mostly this is presentation and docs and web site etc, but it's also about PR and marketing, but most of all it has to be about business strategy. > Thank you for the support and I hope you will help the development of > this project by purchasing a support contract yourself if you use it in > a commercial environment. Please have no doubts on that score. > Again, to reiterate, the product is not -early- in the development > cycle. There are big name customers using it. When you see the name "2.0 > beta 1", it means a beta release in the 2.0 branch which is a newer > version which needs to be tested by our customers. Red Hat also releases > betas for its enterprise distributions.. by no means is their > distribution a premature product. You're once again mistaking my meaning. It's premature in a business sense; the current business infrastructure and the presentation of the product have fallen behind where the product itself has got to. > for your comments, but you have some wrong assumptions from your > perception of the product. I don't believe so, but then I would say that ... :-) Robert _______________________________________________ Openfiler-users mailing list [email protected] https://lists.openfiler.com/mailman/listinfo/openfiler-users
