Slashdot Mirror
Gates Says "A Lot of Work" Ahead In IT Development
An anonymous reader writes "Bill Gates concluded his last Microsoft-associated public appearance in the EU today with comments about the future of IT. The long-time company head said that there's still a lot of work to be done before Information Technology resources truly come into their own. '"There's another side that is how software is allowing people to be more productive at work. It's the empowerment of these people to do their jobs more effectively." Gates also commented on the potential of the Internet, calling it a "huge democratization tool". But Gates said there is still a long road ahead for tech development. "It's come a long way in the last 30 years but we're not even halfway there with building the systems we need to have."'"
That's a revelation? Isn't that what has been promised continually since day one?
The Amiga, Acorn Archimedes, Atari ST and so on were all capable of many of the things that Windows is only recently capable of and yet they were all products of the 1980s. MS has done nothing to advance the state of computing. The resources that are wasted on trying to deal with their proprietary crap would have been better spent elsewhere. Even today with OOXML we are still fighting them while they dig their heels in to slow progress until they are good and ready.
20 years and counting Bill. 20 years. I weep for the state of computing under MS's jackboot.
"I have the attention span of a strobe lit goldfish, please get to the point quickly!"
He's right. And the main reason is the stuff that his company sells.
A consistent picture in every company that I have seen from the inside, with not a single exception: The Unix (or in some places, the mainframe) department is an order of magnitude more professional than the windos group. The Unix servers run reliable (mostly), while the windos network is always a hassle. I've twice replaced the windos infrastructure for a small team with something non-windos (Solaris once, OS X once) and it worked better, with less maintainance, and more useful features.
By now I doubt it's a coincidence, and I've come down from my former arrogance of simply assuming that windos admins are mostly stupid fuckups who couldn't get a job in real IT. If there's one constant in all the cases you see - namely microsoft software - then doubt as you may but the chances are excellent that that's the reason.
I mostly learned that from the one really good windos admin I had the pleasure of working with. He could make things work. But the amount of trouble he had to go to was astonishing. Since then, I'm sure the problem isn't the admins (though they sometimes add to the problem, as many of them are stupid fuckups who couldn't get a job in real IT), but the crap they're forced to work with.
Assorted stuff I do sometimes: Lemuria.org
"We'll be talking about a computer in the desk in the future."
We've had desks for a long time now with for example a big hole in the middle covered with glass and a computer monitor angled up below it. What I don't need is a computer that is also my desk. Why? Just because we can? I want to be able to upgrade those two things independently. Most people have phones in their bedrooms near their beds, but that doesn't mean we need phones built into beds!
"One of the biggest changes will be how you interact with the device. The devices themselves will get a lot smaller,..."
Make the devices as small as you want, but please keep the UI portion of it sized to, oh, I don't know, maybe the operator? (Cell phone "keyboards", I'm looking at you.)
Attention zealots and haters: 00100 00100
You're trying to be sarcastic, but what you said it actually correct. It would be eaiser, because you'd have to write the script once (or not at all, if someone else wrote the script for you), and then all it takes is to run the script whenever you have a problem. Hell, you could write another script to monitor the log files and run your first script automatically.
Automating common tasks is approximately 2 orders of magnitude easier on Linux (particularly on Debian) than on Windows. I'm saying this having done both.
http://outcampaign.org/
I don't know about you, but all the crap coming out of Microsoft Development labs, their new Frameworks, their Best Patterns and Practices, their new Platform (or I think its been rebranded as Windows) SDKs, their Enterprise Application Blocks, their Windows XYZ Foundations, their new Tools to analyze, check and report, their new Features added to every server or operating system.... (I giving up typing at this point), all that just means I have a ton of time I used to spend coding that I now have to spend learning which of the above are acceptable, and which are a pile of donkey balls.
I did this a few years ago, had to investigate what the benefits of Application Server, Biztalk Server, Sharepoint Server, and Commerce Server was. The answer, you'll be unsurprised to hear, was very little to us, but a fair chunk of licencing money to MS (I mean, AppServer was practically a fancy way of configuring network load balancing).
I may be showing my age now, but I wish they would stop releasing new stuff and simply improve the existing stuff. The time and effort they've spent re-implementing Java could have made Vista into a kick-ass super OS that was easy to develop for and would run so efficiently I needn't have bought another 2 gig of RAM and a dual core CPU!!
The reason we build good bridges is because there's only 2 or 3 designs. 1 suspension bridge is pretty much the same as another, when they do try to make something new (eg the Millenium 'Bridge of death' across the Thames which was an upside-down suspension) it wobbled so much they had to close it and debu.. fix it.
The same applies to other engineering structures - skyscrapers, ships, cars, etc. These are all the same pattern and when new ones are built little changes. If there was the same level of "innovation" in engineering as occurs in software, everything would just fall down.
Given the roots and legacy of Windows, there's really not a whole lot MS could have done with Vista to make it a kick-ass super OS short of writing it from scratch (ditch Win32) and force all existing apps to run in a sandbox the way Apple did with OS 9. There's just too much legacy stuff tying Microsoft's hands. Here's a great article comparing what Windows is now with what dBase once was: http://garywiz.typepad.com/trial_by_fire/2006/03/windows_vista_p.html Windows seems to be following a similar demise for similar reasons. What is comes down to is that Microsoft is getting to the point where the best they can offer in a new product is backwards computability with their previous products. Few people care about what NEW software they can run on Vista. Most just want it to run the same software they've always run with a little more flair and perhaps with a little more security.
-matthew
"THERE IS NO JUSTICE, THERE IS ONLY ME." -Death
What do you expect? This isn't necessarily greed on Microsoft's part. It is damage control. They are trying to save the Windows lineage. If too many people keep running XP, Windows as a platform will shrivel up and die. If nothing else, Vista keeps up the illusion of progress and innovation. And Microsoft needs people to buy into that illusion. Hell, I don't even think they really need to buy it. I wouldn't be surprised if Microsoft started giving it away like IE and other products.
"THERE IS NO JUSTICE, THERE IS ONLY ME." -Death
I've switched to OS X just a year ago, dragging several people in my immediate environment with me. From geek to "I bought my first computer two years ago", they all agree that OS X is so much better and ask themselves how they could ever put up with the horror that is the windos GUI.
No, the real truth is that the windos environment is horrible, but it's the environment everyone is familiar with. That means low training costs. Simple as that.
Assorted stuff I do sometimes: Lemuria.org
You imply that the 2 camps are roughly equivalent in quality, and that the unequal bashing is just bias. Not so! On the majority of objective measures, open source is superior. There are the obvious, whole point of libre ones, such as ability to examine and modify the code. And then there are the measures that take a little research to determine, such as which systems are more secure. I've read that CERT has always found more problems with Windows than with Linux. There are plenty of other measures: code quality, performance, robustness, nimbleness, and portability to name a few. Libre OSes are better than Windows in all those categories. And the reason they are better is in part because they are open and cannot be monopolized-- the many eyeballs effect, and the inability of private interests to be the gatekeepers of all progress. Windows does have a few advantages, I'm not denying that. Even when you move to applications, what do we see? Yes, Firefox is better than IE. Pidgin is much better than AIM thanks in part to AIM actually getting worse. OpenOffice vs MSOffice is more a matter of what's important to the user. On the other hand, I read that the GIMP still isn't as good as Photoshop, but it's gaining. And I don't know where gcc stands compared to Visual Studio, but I've read that in the past gcc definitely generated the poorer code but now this is not so clear. Overall, libre is better. I only wonder how long MS can carry on in the face of the massive disadvantages their chosen business methods put them at. For years now we have seen MS resort to unethical methods, and that's the mark of a weak competitor. They are only strong because of their near monopoly position, not because of any inherent superiority to their practices (the ethical practices, that is) or software. They've also made a lot of enemies, not least the previous monopoly computing giant, IBM. Strip away that monopoly, and MS would have to change or die, and they know it. It will be a real shame to see the huge pile of money they've saved up be frittered away year by year in hopeless attempts to maintain the status quo, but activation for XP, WGA, and now Vista seem a clear signal that's the direction they're determined to keep pushing towards.
Those comments that get "modded into oblivion" very likely deserved it for misrepresentation or outright lies. Genuine problems with libre software are fixed right away, or acknowledged. Those that are buried are rare, and often they get forked. Xfree86 and Xorg come to mind on that last.
Your comment seems trollish to me. Consider that maybe the majority of Slashdot has good reasons for believing as we do.
Intellectual Property is a monopolistic, selfish, and defective concept. It is "tyranny over the mind of man"
True, there were several factors at work here.
...", "hm, why does it take so long?", "I think I'm in the wrong workgroup, where do you change that, again?" (hint, it's nowhere even near the network configuration, it's in an entirely different dialog...) ).
True, one of them was that Apple used to be more expensive than PCs. That is no longer true, but it was, for many years.
Other factors are, of course, MS illegal business practices, the original IBM deal, later OEM deals and so forth.
Once dominance was achieved, all the self-perpetuating effects of a lock-in apply. In essence, switching becomes too expensive (in money and/or effort) for most people because you have all your software, all your documents in some proprietary MS format, all the usual stuff.
Finally, Apple does have the full features, they just work slightly differently. For example, OS X supports Kerberos, which alone puts you way ahead of most so-called corporate networks (which are regularly little more than samba shares on drugs). Bonjour is an incredible service for small networks. If you're running a company with 10 or 20 employees, setting up a network based on Apple technology will save you many days if not weeks of work, lots of trouble, and you can do it even if there isn't a geek in your 10-20 people. Oh, it also works.
I have the direct comparison, because I run windos for gaming. Stuff that takes me seconds to do using two OS X machines ("hey I need this file", "ok, I'll copy it into your inbox" (drag & drop some, done)" is a hassle on windos ("I don't see you on the network.", "Try my IP, it's
I've been watching this train wreck since I moved away from DOS 6 to Novell DOS. I've had the misfortune of using every windos version except ME, usually at work. I've seen how Linux came up, had its opportunity, and blundered it. I've seen how Apple resurfaced, and went straight past anything MS has to offer at a speed that's probably over the speed limit. Windos is a pile of steaming crap, from the network layer to the design details of the GUI. The reason it was successful are mostly business strategy, most of them criminal. The reason why it's still successful is summed up in one phrase: Lock-in. Everything else like training costs, costs of converting documents, convenience, software not being available except for windos, is all just manifestations of lock-in.
Assorted stuff I do sometimes: Lemuria.org
Does anyone else find it strange that someone who has built the most successful tech company in history continually delivers the most vacuous, cliched and uninteresting technology predictions of any technology pundit?