Slashdot Mirror
Cringely's 2004 Predictions
somethinghollow writes "Cringely's 2004 Predictions are out, and he makes a very interesting claim concerning Linux: 'The SCO debacle has created a crisis within the Linux community. They pretend that it hasn't, but it has. This will come to a head in 2004 with either the development of a new organizational structure for Linux or the start of its demise. Linux has to grow or die, and the direction it takes will be determined in 2004.' With a claimed 70% successful prediction rate, you at least have to listen..."
"I said that Sun would decline further, generally because of the success of Linux. There is no doubt that this was correct."
"The U.S. IT industry will see some real growth except for Hewlett- Packard and Sun, which will continue their declines."
This guy must be on crack. Sun shipped the most copies of UNIX last year, have lower prices than Dell on the x86 side, supports Linux, created OpenOffice and supported it commercially, and is doing some of the most innovating development of any company (including the open source community, though they aren't a company).
"2) We still won't see a big example of cyber-terrorism simply because nobody has figured out how to actually kill people that way."
What's the fun in killing people? Costing billions of dollars in damage multiple times a year, even a month, seems like a much better option if I was a terrorist. But I'm not a terrorist, I fear caves.
"12) Wal-Mart's entry into the music download business changes everything, and will undoubtedly take the leadership away from Apple."
I think that's a bold statement. Time will tell.
My look at the future of Linux is based on moving from near silicon valley to a hick town in California. Basically, looking at young people in this hick town, I see very little real interest in Linux among the geeks. Yes, there is the "Linux is cool" factor, but I only know a handful of people in the CS department who actually can do anything more with Linux than install it.
The problem with Linux, at this point, is that it is still essentially a free UNIX. Today's geeks do not seem to have the patience to learn the UNIX command line; they are more insterested in the trendy video game of the week and hooking up their new digital camera to their computer (and, yes, Linux has a big problem with USB devices; especially the ones that are really cheap at Fry's). Again, this may be different at better universities than the one I am going to in the hick town I am going to, but UNIX mastery requires both a social group where there is pressure to learn UNIX, and a group of mentors in that group to teach the secrets of the UNIX command line to newbies.
Yes, Linux is making progress to not be a UNIX, by having a much easier to learn GUI for people who do not want to learn UNIX, but the lack of unity (a natural consequence of Linux's open-source nature) and the continuing slow economy that lowers the funds available to help support open source development (a lot of open source development is only possible because the programmers are paid to make the software) both leave Linux behind Microsoft and Apple in terms of GUI quality.
RedHat's decision to no longer make their flagship distribution freely downloadable does not bode well for the future of Linux; I have seen geeks interested in Linux who look at the number of distributions available, can't decide which one to use, and then decide to be interested in the latest video game or whatever instead.
Prediction 1: So no-one knows how to write games with the Cell processor? He could have made the same claim for the PS2's VU units yet I see no lack of PS2 games. There wasn't even a noticeable delay while games programmers learnt to use it!
The rest of (1) is hardly a prediction, unless 'Microsoft will carry on as normal' is an earth shattering revelation.
Actually even the predictions with content too go wrong mostly. You remember all those predictions last time ? like Bruce scheiner had predicted "Major cyber attack" ? Meta group's prediction that Microsoft would come up with linux version of some of their software? I can go and and on with the list of predictions by well knowng guys and firms that went HORRIBLY wrong. Ofcourse some of their "predictions" get hit also(like "there will be windows in the desktop next year also"). Actually I am surprised that at the end/begining of every year slashdot and other news site spend a lot of time for these craps. These predictions are something like a "routinely bullshit" done of every year.
http://www.nasirudheen.blogspot/
I think this could have been predicted for 2003.We've already seen examples of cyber-extortion here (the medical transcriptionist in Pakistan), and I've seen lots of other reports of cyber-extortion attempts here and there, especially in the last 6 or so months. And this is the stuff that is actually reported in the news! I wonder how much of it goes unreported...
People say I'm crazy, I got diamonds on the soles of my shoes...
he makes quite a bit of very spesific sounding but very fucking broad guesses. the kind of guesses you can always say that were true no matter what happened.. and which have no true meaning at all("apple will not sell as many g5's as it hopes"- that could be 'true' no matter how many g5's apple sells)
btw, what fucking new drm schemes did hollywood come up in 2003 that were promptly broken? what?? anybody??
world was created 5 seconds before this post as it is.
1. The PS2's VU was pretty hard to write software for, but who is winning the market right now?
2. Nintendo have not announced what proecssor they are using, so how can it be the Cell processor? Who said that Nintendo, Sony and Microsoft were releasing now consoles this year? At this time, they are all gunning for 2005.
3. Apple are not going to release flash iPod's, instead they are going to release HD based iPod's with 2 or 4gb capacity. This is a solid rumor.
4. Apple have made no announcement of how many G5's they want to sell, so anything is not what they are hoping.
5. Chances are the G6 will be released next year as the Power5 is being released next year.
6. Linux die? How? It's not a company, its a conglomerate of programmers. It's marketshare is rising, not falling. Case in point, OSS such as Apache is only growing in popularity.
7. How is Microsoft continuing on their normal ways a prediction? It's a fact.
8. Walmart are going to have some serious issues with their online music store simply because its not easy to use. I agree that Apple at this rate will not be in the lead though.
9. The Burst case is interesting, but I can't see Apple and Real being punished if Microsoft loses/or buys Burst.
All up a rather silly set of predictions that is all too vague or missing facts. I can see why he gets 70-80% success.
I think what he means is either more corprate structure or less corporate structure.
For example, currently driver support in linux is horribly broken. Worse yet, it is horribly broken *by design*. The kernel developers refuse to freeze a driver api for the kernel (like MS did with win95/98/me and win2k/xp) or make a DDK (to my knowledge) instead changing the api every major release and sometimes on point releases. They have a point -- doing so encourages the development of open source drivers instead of binary only drivers. Unfortunatley this is one area in which corporate interests might superceede the interests of the individual kernel developers. Possible.
Another area we might see is with regards to linux on the desktop. Progressing, yes, but not as fast as it might if - say - sun or ibm decided to make a serious commitment to linux and either extend kde/gnome/x or replace them entirely. If they come up with a desktop replacement that is far better than the current offerings they will win by default, thus placing the future of linux on the desktop in corporate hands, at least for a short while.
I think this is what cringley was getting at -- changing linux to reflect the fact that alot of the development is funded by major corporations now instead of hackers working on their free time.
Exactly. It is entirely true that Linux, as an OS (kernel, whatever; it's a pointless argument) need not show the kind of year-over-year growth that proprietary software vendors like to see from their products -- there's no Linux accounting department that will axe the project if its ROI fails to meet expectations. But: there is a difference between Linux, the cool open-source geek OS, and Linux, the business-computing phenomenon that is the best chance of toppling Microsoft from its throne. The first will survive, in some form, whatever else happens. The second is in a very delicate place right now. What I hope to see happen is that with backing from Big Blue, Linux continues to gain mainstream acceptance and eventually becomes one of the default choices for corporate use everywhere from the desktop to the mainframe. But that won't happen unless it makes a few more major steps in the propaganda war. Right now it's seen as a respectable alternative ... but "respectable alternative" and "mainstream" are not the same thing.
The correlation between ignorance of statistics and using "correlation is not causation" as an argument is close to 1.
The Linux kernel development already has a tried and true tested organization. Linus Torvalls and crew operating as benevolent dictators, totally open to public scrutiny, with no abolute power to dictate what additional patches the distributions and developers end up using. It has worked and continues to work very well.
All the contributions and development are traceable though both the Bit-tracker/CVS logs and the mailing lists, which makes everything available to public scrutiny for everyone, include those who are publicly defending Linux.
Throughout 2003, the SCO Group's so-called evidence and legal theories have fallen into disrepute though the rediscovery of the combination of the terms of the GNU General Public License and the open development process of both the Linux Kernel and even UNIX itself.
The weight of the historical evidence, including the active participation of both old SCO and Caldera executives and employees in the development and promotion of Linux, tips the scales of justice heavily in favor of IBM, RedHat and Linux end users. In fact the weight of evidence effectively chucks SCO legal position off the scales, out the window and over the cliff like a cartoon catapult.
While each new Darl McBrides threat and new David Boies partners legal theory look impressive at first glance, in practice they are about effective as attaching a giant anvil to a biplane to catch a pigeon.
See http://www.tibonia.com/Dmeg1.htm
The McBride and Bois Show to stop the Penquin is becoming a joke ...
It would still only be illegal inside USA if it came to that, so I don't the rest of the world would notice much change.
And at the moment, it seems that Linux has the fastest growth outside USA.
"As for old fashioned spam, it will continue to cram our inboxes, making a good business for third-party anti-spam products and services while making e-mail pretty much useless for reliable communication. Microsoft will see opportunity here and propose new protocols to replace SMTP and POP3."
Why replace POP3 (and IMAP)? These work fine and are completely separate of the SMTP delivery engine. The smart thing to do would be to replace SMTP MTAs with something that does server-to-server authentication, and leave POP3 and IMAP for the MUAs.
--
Internet Explorer (n): Another bug -- that is, a feature that can't be turned off -- in Windows.
Grow is too much of a corporate buzzword, I'd replace it it with "find a goal". The original goal of a free, stable and usable Unix workalike has been accomplished and then some, with ports to various architectures large and small.
So what's next? Is fine tuning the VM and implementing yet another filesystem "it"? Or should there be another goal that takes it somewhere else, like a Desktop Linux initiative or something? I'm guessing that's the growth he meant, not in the "get bigger" sense, although I'd imagine he was thinking of a revised goal that accomplished a bigger user/installation base, too.
Pick a low quality (costly) product. It comes under pressure from a free high quality product. The low quality (costly) product comes under pressure. A 3rd grade kid could draw that line of reasoning.
You would think that, though my managers still think OSS isn't ready for prime time. Most don't realize that Apache, Tomcat, and a variety of other pieces used on our 'flagship product' (don't ask...) are OSS! The looks and speachlessness that occurs when I point this out is amazing. They really don't get it...and these are the smart people.
A firewall can not protect you from yourself. Turn off what you do not need. Do not use the firewall to do your work.
OK, so he stopped predicting the sun would rise tomorrow and got on with some original thinking. And failed, though it was a nice idea.
Actually, the process has stalled and was nowhere as fast as people thought, but China is still pretty much moving in that direction. So it is too early to say. So the wrong part here is that he sais that his predicttion failed. Also it is not MIPS(tm). Some parts of the instruction set and paten encumbered parts of the ISA are missing
Baker's Law: Misery no longer loves company. Nowadays it insists on it
http://www.sigsegv.cx/
This is just an old trick, popularized by horoscope and fortune cookie writers, applied to the computer industry. There's almost no conceivable course of events during the next year (especially given the certainty of more SCO-related headlines, at least on Slashdot) that couldn't be described as at least provisionally consistent with this prediction. If Cringely provided some analysis or insight, I guess there could have been an interesting point here. But it's really just a throw-away.
As an exercise, review the years to date for Linux. Each year has seen some significant movement towards both growth and death. Lately there have been interesting maintainer changes, which would probably qualify as changes in organizational structure. It's hard to remember a significant period of time during which Linux hasn't confronted something that could be loosely described as a do-or-die challenge. The same could probably be said for most if not all organizations in the tech industry. If Cringely really wanted to make a non-obvious prediction, and this was the best he could come up with, he probably just didn't give it much thought.
When they do that, they would lose probably more than half of their revenue and Bill Gates' and Ballmer's stock would become a lot less worth than it is now.
I predict Microsoft will continue to do selective discounts, or to put it in another way: To offer great rebates to possible defectors while ripping off their loyal userbase.
Sure. Just like Motorola did. China is just like Japan in that they don't want to actually open their markets to outsiders. They want to milk off all the technology they can (by either waving dollar signs in front of naive CEO's or simply stealing what they need) and then selling it back to us to maintain the trade deficit. No, I suspect Sun has been blindsided and MY prediction is that the Chinese will take them for a ride.
The higher the technology, the sharper that two-edged sword.
In the Linux 2.6.0 kernel, under the "drivers" directory, there are currently 2218 ".c" files :
/usr/src/sys/kernel/linux-2.6.0/drivers
.c source files. Using that assumption, that calculates out to 1109 open source drivers in the current Linux kernel. That is a lot !
.c file, and in a number of cases, multiple devices from different vendors may be supported by a single .c file, as the they have used underlying chipsets from the same third party. So the current Linux kernel would support say 3/4 of that 2218 figure, or 1663 devices, which all have fully open source drivers. The figure is probably significantly higher, possibly even greater than the 2218, but we'll stick to 1663 for this example.
--
> pwd
> find . -name "*.c" | wc -l
2218
>
--
Being very conservative, let's say each hardware device supported by Linux requires two
Of course, most drivers are only contained within one
Let's divide that 1663 devices by 10 to get the number of manufacturing vendors. Of course, the figure is likely to be much lower than 10 devices per vendor, probably 5 or less, maybe with an average of 2. However, let's assume 10. That indicates that 166 vendors have 'got' the open source idea, and published their programming specs, or even contributed GPL licensed drivers.
Linux driver support may appear to be broken because the two major graphics card vendors don't publish programming specifications. However, 2 vendors holding out, compared at least 150 or more who get open source, doesn't make Linux driver support broken.
The Linux development model has succeeded, as has the Linux philosophy on driver development. It is only the few, high-profile "bad apples" that make it appear to have not been successful.
The Internet's nature is peer to peer - 20050301_cs_profs.pdf
I don't find this all that likely. Maybe that's what he SHOULD HAVE BEEN getting at... but you're giving him too much credit.
The World Wide Web is dying. Soon, we shall have only the Internet.
With all due respect I think some people have a tendency to assume that since Sco's claim is likely to be false that is the end of the issue - its not.
As a thought experiment assume that suddenly Sco do come up with some code. So we can just rip it out and replace it and everythings back to normal right? Of course not. Imagine I break into your house and steal your stereo. Six months down the line you come round my house and find it, so I give you it back and go and buy a new one. Would that make everything alright - of course not.
The one argument of Sco's that I do not see being fully addressed concerns the lack of oversight. Sco argue that open source development is fundamentally flawed because of a lack of it, or a least a formal structure to deal with it - and this may well be the point Cringely is pushing at. It would only take I line of code for there to be a argument to answer to, and even if there isn't that argument might still have validity.
So to continue the experiment, what if they do come up with goods? what if they don't but companies start to feel the chill? The stage is set for somebody to come along and say Linux is a great thing, but its a little too anarchic, what it needs is a formal structure. (Guess what? someone might just have said it). At this point the stage is set for large vested interests to come in and say we've got the money, we've got the people, the most efficient thing to do is set up a top down structure(that just happens to have us in control).
At this point the free and open nature of the process is under threat - it isn't *just* about the code - it never is - in this field or any other. In the worse case scenario Linux could effectively become just another coporate product - but one backed up by unpaid labour - and how long would it last. The GPL does provide a good gurantee - you can always fork (as long as the GPL holds) - but this could just translate to death by a thousand cuts.
Don't think that popular open movements can't be taken over by self-interest masquerading as efficiency - outside computing it has happened time, and time again.People need to start thinking seriously about these issues. We need to consider how we can create structures that can answer these issues without compromising the open nature of the free software process. Don't think of this as a dumb prediction, think of it as a wake up call
Paul M
"There are no innocent bystanders. What where they doing there in the first place"
William S Burroughs
Purely regarding financial methodology, rather than Sun in specifiecs:
Profit and loss accounts refer to the past, not the future. If a company did restructuring and trimmed poor divisions it may make a short term loss but would benefit in the long term.
The price of a stock refers purely and simply to its supply and demand, i.e. where people are happy to buy the stock and where people are happy to sell.
So the price if a function of many agents' opinions of the future price and earnings of the said company. This is inclusive of technical trading, but I won't complicate going there.
The stock price does matter as it is the quantification of qualitive opinions regarding the company. If you don't think a company is priced well, put a massive short position on the company and report back sometime in the future. I look forward to seeing you on it.
karma karma karma karma karma chameleon, you come and go, you come and go.
I know I'm just stating the obvious here, but it seems worth mentioning, in case people really believe Cringely's "the sky is falling" claim regarding SCO and Linux.
First, there is Linux the kernel, and there is Linux the full featured operating system. The only thing in any danger is Linux the kernel, which is just a part of Linux the operating system. So, right off the bat, the potential danger is localized.
Second, once SCO is compelled by the courts to reveal the allegedly infringing code - if there is any - the Linux community will quickly replace it, and Linux will be back on track.
If, for some reason, replacing the allegedly infringing code is not possible, there are other kernels to turn to, including but not limited to the excellent BSD kernel (Free, Net, Open).
This SCO nonsense is just good entertainment for us, and a foolish money sink for companies like IBM that have to put up with SCO's obnoxiousness.
There's nothing to see here. Move on.
-Teckla
We're talking apples and oranges here. I'm quite fond of Knoppix (and have hopes for Gnoppix as well); I just used it to convert my wife's Windows laptop to GNU/Linux/KDE. Price is not the issue in this debate -- attendance to user needs is.
For example, I miss the graphing abilities of Excel; OpenCalc and Gnumeric simply don't compare. The usual "free" response is to use gnuplot or some other outside package, or to modify Gnumeric myself. I'm certainly capable of doing both, but I haven't the time; I'd prefer a "we've haven't had time to do that yet" over the more typical "we don't give a damn, quit complaining."
As you point out, Microsoft is equally guilty of indifference to users, given that they are far more concerned with sales and marketing than with whether their software runs reliably or well. Sadly, users get screwed by both sides of the software spectrum, albeit for different reasons.
All about me
If I understand it correctly, Wal-Mart's core competency is using its size to negotiate down the costs from distributors and producers. I suspect that if Wal-Mart is offering their music for less, they are paying less to the record companies. The nexus for the agreement may very well be the more restrictive nature of the consumer's use of the Wal-Mart music files. There may even be co-op money for Wal-Mart from Microsoft for promoting the WMA format.
As this is the entertainment business, I'll just point out that when HBO started to look like it could be a success, other companies started movie channels. And within a few years all the major film studios had exclusive deals with the channels for the right to be the first channel to show the studio's releases. Something similar happened when others saw MTV's success and they followed behind. Deals for 90 day (IIRC) exclusivity proliferated between the channels and the record companies. I wouldn't be surprised if exclusivity doesn't begin to appear this year, and not to Apple's advantage.
Cringley makes one of the classic tech-punditry blunders, which is to confuse Linux with an operating system while simultaneously confusing it with a religious movment and/or trade association.
It's none of the above, of course. It's a free software kernal, rolled into many operating systems like Red Hat and Debian, but still just a kernel. Pretty much useless by itself, unlike Free, Net and OpenBSD, which are top-to-bottom OS projects, with a central organizational structure that takes care of everything a user could want or need in their Unix system.
Free Software/Open Source has not one, but two religious movment/trade associations, complete with Famous and Glamorous grand high pooh-bah charismatic heads. Richard Stallman on the one side, and Bruce Perens on the other. Both men and their organizations are pretty much ignored by everyone involved with Linux, save to incorporate their software into the Linux-based OS projects or to toss obscene amounts of cash at them to help them kick Microsoft out of the datacenter. Overall, they're mostly just good for really entertaining flame wars.
Linux will continue to grow unchecked because there is no organizational structure. People are free to take and use the kernel however they see fit, so long as they share the source code to any modifications, so it will wind up in spacecraft microcontrollers and kilo-processor supercomputers, wrapped in the software needed to get the job done.
Linux-based desktop operating systems will put in more effort to be interoperable with each other, though it's unlikely they'll all get together and decide to have someone be their collective boss. That's not neccesarily a bad thing, and coprorate customers will be more comfortable knowing that the organizational structure in charge of their Linux-based OS is "IBM" or "Red Hat" rather than a nebulous organization of hippies and geeks... gives 'em someone to sue if it all goes wrong.
SoupIsGood Food
As a member of the Linux community...
Ever noticed that those who have to say it, aren't?
Pros: "I know a little about computers."
Also-rans: "I'm a computer expert."
Pros: "We'll do our best."
Also-rans: "We deliver quality."
Pros: "I'm OS neutral." (though would probably recommend specific OS for specific job)
Also-rans: "Linux is like a god."
Pros: "Life always picks up and goes on..."
Also-rans: "Linux could be threatened! Everybody should worry about it!"
I'll bet that:
90% or more of what we worry about, in life, doesn't happen.
90% or more of what we hope or dream about, and really work at, happens.
I think Walmart's core competency is not so simple as that. They are primarily a brick and mortar store. Near as I can tell, they haven't gotten rid of their online competition.
Just anecdotally, I have heard that Walmart's selection of music priced under 99/song is quite limited. Other complaints I have heard is that the shopping process is complicated, and that the DRM terms are inconsistent and confusing. So far, they don't seem to be executing so well.
I don't think your cable TV examples prove your point. HBO is still going strong. In fact, it might still be considered an industry leader. Exclusive deals weren't new with HBO or cable either.
Exclusivity has appeared already. iTMS has tracks from various artists that are exclusives to the store (I'm sure you can find them on the p2p networks soon after, of course). As I said, these sorts of promotional deals are nothing new. (I recently purchased an exclusive single from iTMS: The White Stripes "Black Math".)
Walmart succeeds best when selling commodity items, as you mentioned. True, they are big and they can negotiate based on that, but currently Apple has the highest volume in download sales, which also brings something to the table.
Another thing to keep in mind is that while the entertainment industry is mostly ruled by bean counters and lawyers, it is still an ego driven and bullshit fueled business. Who do you think has a better chance of successfully bullshitting someone, Sam or Steve?*
*If you've never been within the radius of the reality distortion field, you've got to experience it. Come to MWSF or the WWDC later this year. If we meet up, I'll buy you a glass of kool aid.
Anyway, I'm not arguing that Walmart and MS don't present a threat to iTMS or Netflix. I agree with you there. I just don't think it's a foregone conclussion. First-to-market advantage is a big deal, and first-to-market-done-right is a huge deal.
It's not offtopic, dumbass. It's orthogonal.
The reason to Panic is not that Linux may be declared illegal itself, but that the Open Source development process behind Linux may be declared to be "unable to ensure the legality/authenticity of the produced IP".
This would not just require replacing the Linux kernel. It would make any software based on the Bazaar methodology potentially illegal and kill both commercial creation and commercial/professional use of such software.