Newer open source software doomed in enterprise says SAP - Computer Business Review

I was fuming after reading about Dovecot Falling Down today. UW-Imap has been bug-report-free for longer than Dovecot (it's a mail server; can't you tell by the name?) has even been on the planet, and now Dovecot hasn't been bug-free for as long as UW-Imap. The kids - who were all born after sendmail, I should mention - seem to love it and the distros will probably still include this provably-worse software in their distros.

And Dovecot isn't even out of beta yet, and it's in Enterprise offerings. You catch that one? Enterprise. You know, the stuff that companies want to sell us, which can apparently run our stock markets, warships and nuclear research labs.

I'm not saying that national defense will hinge upon our ability to send mail on-time and securely - although it may soon - and maybe that kind of organization is the upper end of Enterprise, but the idea that this kind of software package will be part of what people want to sell us for that kind of scale of business is really unfortunate.

Don't get me wrong, O thin-skinned developers of Dovecot who cannot take the written assertion that your beta product is not fit for military use, I'm sure it's a good product and you're all very proud of it. Let's hug and let me go on.

So I was fuming about OS vendors using such new and untested products in their offerings, forsaking those with a track record of any kind, and I came upon this article. Same day, kids.

Newer open source software doomed in enterprise says SAP - Computer Business Review:

As CIOs look to consolidate their IT operations, relatively immature open source software has little chance of surviving in the enterprise, said an SAP AG executive during a speech at the Open Source Business Conference in San Francisco.
Advertisement

That consolidation wave has already begun, both with vendors and among enterprise customers that seek lower operating costs and complexity, said Peter Graf, SAP executive vice president of solution marketing.

'The question for customers is, Is this [software] stable enough to bet the farm on? They cannot afford to risk their companies on a technology that may not be stable enough to carry their business in the long run,' he said.

"So there's a lot of concern on the part of the buyers, now, because they don't want to bet their business on something which may crash.

It's all kinda vindicating, in a way, that someone out there agrees with me: choose stuff that you know works, because you can see the day where it broke and the day where they fixed it. Let these upstart projects cook for ten years and see how well they fare -- I'm looking at you, postfix. I'm thinking that, if they even do remotely better, it'll have something to do with population increases and more Internet use. And that's it.

We're seeing Microsoft, for instance, taking the piss out of Linux, these days, pointing to the bug reports on CVE (linked above) and whatnot, and saying that Linux has a lot more bugs than Windows, and just look at the stats. The sad truth is that those stats aren't wrong. Look at them. The amount of crap that's bundled into these Linux installations, just so they have the features they want, is really starting to make them look bad because of all the bugs being found in these untested backyard projects.

I think it's time they took a long, hard look at products with track records, and decide what they want to include in the next run.