New Red Hat CEO checks open source claims
Not my bag, baby
The news that the operations chief from a major US airline, Jim Whitehurst from Delta, is taking over at Red Hat from Matthew Szulik is a further sign of the growing legitimization of open source and Linux in the eyes of corporate, mainstream America. It underscores how the “suits to sandals” ratio in the open source and Linux movement sliding further towards the suits.
Whitehurst’s blue-chip background at the $16-billion-a-year Delta and the fact he’s an executive lured from outside of the IT industry rather than one who simply swapped one tech industry management job for another underscores the belief in open source as a business.
Recent research from Saugatech found 10 per cent of software used around the world is open source and that enterprises are increasingly taking open source seriously.
And even that perennial chestnut - the growth of Linux on the desktop - seems at last to be coming true. This month's stats from the W3C shows that the percentage of Linux desktops has grown to 1.77 per cent. This might seem ridiculously small - but six months ago it was only 1.25 per cent.
It's not only Linux that is growing. With Firefox 3 looming, W3C also shows Firefox is making serious inroads into Microsoft's dominance of the browser market. The latest version of Firefox now ranks second behind Internet Explorer 6.0 with almost 21 per cent of the market. If you stir in the effect of a mere handful open source packages such as Audacity in audio processing, Gimp in image processing and OpenOffice, the penetration of open source software in 2007 is striking.
It is worth re-iterating here that open source software is not always "free". Indeed, some of the more recognized and successful open source packages such as Red Hat's Linux while not free are less expensive than proprietary equivalents - no names mentioned.
It's been a while coming, but as noted in David Wheeler's comprehensive paper on the growth of open source, it has been building up for several years. As is often the case, growing interest in the consumer market is echoed by growing interest in the enterprise market.
The key to the success of open source is its maturity. Not only is the code for many key open source products now stable, but the processes for maintaining and developing it are also formalized and efficient. Gone are the days of ad hoc changes and dodgy code. Furthermore, open source products are generally becoming as easy to use as expensive proprietary products while thoughtful, altruistic people have populated the web with free guides and tutorials in abundance.
Although they have swooped on open source as the next big thing, suits like Whitehurst could be in for a disappointment. The open source movement is generating a wide range of excellent software that is available for no more than the effort needed to learn how to use it. This will make it hard to sell any products - including some open source software.
But it is not only the suits that need to do some hard thinking. The success of the open source movement also raises serious questions about the future economics of software development. If, as seems likely, that in the not-too-distant future all software will grow from open source initiatives, who will actually pay for programmers?
Whitehurst himself stepped into Red Hat from a struggling sector. US airlines are also trying to find ways of making money against the competition while their customers have more choice than ever before. In April, Delta itself emerged from restructuring under Chapter 11 bankruptcy protection. While Red Hat has seen phenomenal growth, the company faces its own challenges from making the JBoss unit live up to pre-acquisition promises and fending off Oracle CEO Larry Ellison’s low-priced support network to standing up to next year's Windows Server 2008 and Microsoft’s on going claims of Linux and open source patent violation.
One thing is clear. The richest people in the world in 20 year's time will not have built their fortunes from making software, which will come as a hard shock to people like Whitehurst who might have hoped to retire rich on open source’s fortunes. If Microsoft's Bill Gates or Larry Ellison of Oracle were in their early 20s today, they would be looking for other ways to make their money.®
Additional reporting by Gavin Clarke
RE: This is broken in the release version
Not take over, increase the installed base in a slowly but surely manner...yes.
It's not black and white, there are different shades of grey. Not that Windows has to vanish for Linux or Mac to surface.
Or are we all still driving Black coloured Ford T's nowadays?
Well, my point is that you CAN open that pdf document but it so happens that not with the default application in certain distro (thanks for pointing this out, I will change the settings for that type of file). You might need to do a right-click and then another click but that would render your .pdf as intended. No need to generalise or put it in a way that sounds like "Linux won't open PDF". I take it you knew other software would open that file correctly (KGhostview and KPDF for example) but you didn't bother mentioning it...hence disinformation. Or at the least not a constructive attitude which helps nobody (well maybe M$).
Before I go on into your next paragraph I have to mention though that Adobe Reader 8 in Windows opened that file flawlessly fast too, so no point in doing benchmarks there (at least in the PC I tried, perhaps benchmarking this in older machines would make a point).
Regarding breakness I have been using the debian distro on the Desktop for nearly a year on a E6300 and I have yet to get it thoroughly broken. Sometimes I might have run into dodgy stuff by installing conflictive software from non-debian repositories, but it was very easy to do "apt-get remove nameofproblem" and then either installing another version or look for a replacement.
It is a different way of doing things, but I feel as it is much easier and quicker to tell the computer "install this" or "remove this" than all the parade that Windows makes you go through whenever you want to do samely.
And many a times you can also find yourself with broken Windows due to dodgy hardware/drivers, welcoming you with a BSOD. Yes this can be solved, but you do have to take your time to boot in safe mode and do the deed and restart again hoping for the best.
In the worst case scenario the whole thing will be non-workable and the only viable solution (or the one most widely accepted) is to reinstall the whole thing, which will then take you a good half a day.
You can use Ghost too, but does the average user know how to operate that?
I haven't the time right now to pick on each of these: "poor documentation, unfixed bugs, badly designed code, imitations of other software, poor quality user interfaces" and I don't think it will be worth it as I doubt there's people looking at this that could benefit from this. To make it short:
- Barely anybody makes any proper documentation nowadays and I still have to find people that actually reads it when there is. The software in fact should be engineered for the average user and software not to need documentation (that's why we decided to use icons and stuff after all isn't it?)
- Everybody has bugs (and some of them take forever to fix them, see http://www.theregister.co.uk/2007/12/20/more_vista_copying_problems/).
- I dread to think what the code will be like in some applications that have been building on top of the same code for years/decades where the programmers/tools have also been changing.
- Imitation as in the same software but now free of charge, and perhaps faster. Sure that at times it won't be as polished or fancy to the eye...but we also did switch from 35mm to digital photography because it was cheaper and quicker.
- And for dessert I'll mention another Windows parade about the user interface...clearing the recently used documents list. In XP it took me at least 3 times the amount of clicks and a lot more mouse travelling than it did in Gnome. I hope Vista helps that one too, because we should be heading for usability and getting the user rid of nonsense so that he can effectively get on with his/her life.
BKB, OpenOffice.org and reality
First, a bit of background: I've been "in the process" of switching myself and clients from Microsoft Office to OpenOffice over the last two years. I've been running exclusively OpenOffice on my own systems for nearly a year now. I support roughly 3,700 users in 71 companies in five Asian countries who are, with fewer than a dozen (users) as exceptions, running MS Office-free systems. These include banks, government contractors, electronics firms, the management offices of a grocery chain you've probably shopped in if you live in Singapore or Malaysia, and so on.
Once the installs were done and the training completed, actual second-level user support takes approximately ten hours per week. At least three client organizations have cost and productivity metrics that say they're saving the entire cost of the conversion and training in, at most, *every six and a half business days*.
The grocery chain plan on hanging onto some of their XP licenses until late in 2008 but are aggressively pursuing alternatives to downgrading to Vista. Their Vista pilots were, in a word, unsuccessful... partially because they were using their existing, year-old, XP-class systems, partially because the negative response to UAC was so strong (despite generally positive user experiences for Mac and Linux security in the same organization).
OpenOffice is saving real companies real money, and the good ones are passing along those savings to the customers through lower prices, improved services, or some mix of the two. It's where sendmail was five years ago - most of the people who kvetch about it the most have the least real knowledge of it.
Now, on the other hand, if you'd chosen to pick on Firefox 2.0.0.???, I wouldn't have been too ticked. In 30 years of IT, I have seen very few programs that so eagerly hand incredible amounts of rope to the user, with which to hang himself. If I'm not careful, I can wind up with 4-5 open instances of Firefox, with 25-30 tabs between them...at some point 2.0 hits an internal brick wall and falls over. At least it can resume on restart to where it was within a few seconds before the crash. I have yet to see this with Opera 9. I devoutly hope that Firefox 3 fixes the memory management problems (as first reports of the beta indicate that it does).
The point of this ramble is twofold: there are dogs and winners in the free/open source software world, just as there are dogs and winners in the commercial/proprietary world - and your list of either will differ from mine because your needs and usage are different. Secondly, as with all topics, it greatly improves one's credibility if one refrains from making overly broad generalizations without at least the appearance of supporting detail to back it up - especially in an arena (such as El Reg) where doing so is likely to trample on the experiences/insight/intelligence/faith of others. Blind faith has never been particularly effective at combating blind (other) faith; much less effective against logic and/or reason.
We now return you to your regularly scheduled blather.