Open source technology, or products — choose, dude
January 9, 2008
Last fall I reported that an HP manager cast doubts about the mission-critical stature of some open source software. David Claypool said in an HP Technology Forum talk that an open source solution better have evolved beyond technology to product status, if you want to consider using it. Evolve here apparently means being part of a company's product line, from what I understand.
Okay, October of 2006 seems a long time ago, but Claypool said he just located the blog report. He sent me an extensive rebuttal, attached to the original post. I love comments on blog entries, even when they tell me, like this HP Product Manager in Technical Marketing did, "I think you took my comments out of context and you missed the point."
Claypool has a lot more to say about his talk and how he believes I misunderstood it. "The important thing," he says, "is to distinguish between open source technologies and open source products. When it comes to a bet-your-business decision to adopt a product, no one in their right mind should plan on doing an Internet search and download a zero-dot-whatever release of something an individual has crafted."
Using that zero-dot stuff is pretty foolish, to be sure. But the cartoon above spoke to this creeping commercial elitism I see entering the open source world. Apparently, if your open source solution doesn't have a company attached to it, and collecting support fees and managing your updates, you're riding bareback.
I think perhaps David doesn't know much about the HP 3000 community, which only got its first working Web server once an individual, the brilliant Mark Bixby, ported Apache to MPE/iX. HP later took it on as a supported product, but millions of sites use Apache in commercial situations without a care for what company the solution gets sold by. Apache is well tested, yes. But in 12 months, HP won't be patching it anymore.
Missing the point? People are choosing open source for several reasons, but one significant motive in the 3000 market is a way to avoid a trap by a vendor. Forgive this community if they don't feel safe in the arms of a single supplier. I know of one vendor who's active in migrating 3000 customers, and they don't want to use much that isn't open source. Products, yes, but not always offered by companies.
Claypool makes a good point about using tested products in mission-critical enterprises. Typical HP 3000 shops are risk-averse and late adopters, so they test like a fish draws water. What’s more, HP has often given the community open source products with no support except the open source community. Python, now Java, the list goes on and on. Some were candidates for support, but never made it. Others were never considered. Still, HP does a good thing by making the latest software available on its Jazz Web site. Disclaimers abound, sure. But sites use this open source to run their companies.
Company-supported open source products aren't available for 3000 shops now, and you might argue that's a reason to migrate away from the platform. Technologies are just about all that the staying-on-the-platform 3000 sites have left in open source. I don't think I misunderstood when Claypool said:
“I’m not going to trust my business to some hacker in Denmark who’s got a ring in his nose and is awake when I’m asleep and asleep when I’m awake.”
And thus, the cute cartoon above, complete with beanbag chair. One wonders what developers outside North America wear, or when they work. But they do develop for HP, even while you're asleep.
This kind message paints independent work from in the open source community with an insinuating brush. The vendor is of more than one mind on this subject. HP believes its 3000 customers are capable of keeping up open source solutions, or the company wouldn't have put out an open source porting paper last month all about updating Samba. That's not bedtime reading for a home computing hobbyist. It's doing your own support, using the power of the worldwide open source bazaar, instead of a commercial cathedral. And frankly, the value-for-money measure of paid support for open source products varies widely.
I still believe Claypool's point sounded like only a company could provide an open source product for mission-critical user. There's this sound of vendors using source code to slide toward profit, even while they capture customers who want independence. Lots of companies bet their business on open source products not offered by companies. There was a time when the Contributed Software Library, the 1980s version of open source, powered some operations at companies like Boeing. Just like many customers, such as air carriers like WestJet, still use a product called the HP 3000 — despite HP’s declining-to-zero-someday support.
I'll give Claypool his last word, to be fair. Watch for his warning that using raw code means that you have to be self-reliant — a skill which much of this community has been polishing since 2002.
Certainly, it’s possible and may even be prudent for some to download and run the bits from a raw open source project — but it’s incumbent upon the adopter to understand the commitment to self-reliance that’s being made if it’s being used in any operational or revenue-producing capacity.