Slashdot Mirror
More Linux Predictions for 2004
An anonymous reader writes "Experts, shmexperts - it's time for the Linux community's own predictions, felt the editors of LinuxWorld Magazine. Prognostications in their Jan 2004 round-up cover media players ('turning your phone into an iPod will be hot by the end of 2004'), IPOs ('Of course, LinuxCertified, Inc'), and MS ('Microsoft will start an intensive campaign to promote their Longhorn technology as Linux standards compliant') - that last is one from Samba's John Terpstra." The original story was back in November.
Sun and IBM will be considered the biggest Linux players by the end of 2004, and that Linux will be installed on Mac like numbers of corporate desktops (corporate not techy).
I also predict the return of thin-clients to the corporate environment, especially in large outsourcing contracts.
An Eye for an Eye will make the whole world blind - Gandhi
Linux, for me, peaked in usability/reliability in 1999. It's still quite useful, but I began experiencing many more compatibility problems since that point.
I have a video card whose driver is closed. I've got multiple peripherals that are only partially implemented because manufacturers for some reason are reluctant to release information to developers. It's great as-is, don't get me wrong, but participating on the Internet has gotten much harder as everybody decides to go proprietary and tug in different directions.
For example, Flash runs slower on Linux; so slow that it causes the sound to go out of sync (related bug that also seems to bite some Windows installs: this applet and those coded like it have audio that is too quiet). Java is still a real pain to get working right. Maybe the greatest thing that's happened this year is Mozilla/Firebird, but I'm running it without add-ons!
I believe only great things are to come, what with Linux having reached 2.6.0, and greatly appreciate all the developers have done for it. Now, I think it'd be nice if others began to support it.
Try not. Do or do not, there is no try.
-- Dr. Spock, stardate 2822-3.
Seen Google's new logo?
I think the general uncertainty in the market will prevent 2004 from being any more the Year of Linux than 2003 was. Not to say there wasn't any growth in the last year or that there won't be more this next year.
I certainly think that Microsoft sending out numerous free copies of Small Business Server 2003 shows that they are taking Linux much more seriously than previously. And I think when we hit 2005 and companies have to make a big decision either way that if the Linux offerings by then for the small shop and desktop have improved their UIs so that virtually anyone can setup Linux on their current machines as easily as or more easily than a Longhorn upgrade, THEN you will see the mass migration.
FWIW...
Predicted that some 13 years ago.
Where is that guy who'd die defending what I had to say when I need him?
As far as hotkeys, why would you want to standardize them? I can define any key to do what I want currently with my distro (SUSE). Different people work in different ways. Why restrict them to what you think should be standard.
Your "frequent tasks" comment doesn'r provide any examples, but you could look back to hotkeys to provide solutions.
tar -zxvf is now automated by a tool called 'swaret', which is an apt-like utility, that downloads/decompresses the tarball, and then work's out dependencies, and download's anything you need. There is pkgtool for tarballs in the slackware format, however it doesnt dependency check.
-Adam
#!/bin/csh cat $0
I think we're on the brink of the collapse of Microsoft's office suite monopoly. There's a lot less lock-in with office than there is with windows, so it's much easier for people to switch to open office.
Microsoft's pricing and online activation system has already pretty much removed office from consumer pc's. People who used to take cds home from work are doing without, and it's only a matter of time until the word about open office gets out. I'm not claiming that open office is as good as microsoft office, but it's good enough, I think.
I think that microsoft is making one of the biggest mistakes in its history in the way it prices office. The strategy seems to be aimed, as near as I can tell, at keeping corporate revenues high while allowing MS to cut prices for low end consumer machines.
A corporate workstation with xp pro and office pro pays microsoft almost 3x what a consumer user with xp home and works pays. I don't think that reflects costs or utility to the customer.
The most useful part of what people pay microsoft for comes from xp home -- it gives you the ability to run the huge library of windows software, access to the huge array of hardware device drivers, and core networking tools. What you get, for the buck, from jumping to xp pro or adding office on to the back, provides a lot less utility for each dollar spent.
If you decide that the corporate market can bear substantially higher prices than the consumer market, and if you notice that the main differences between a corporate user and a home user is office, then loading up the costs on the office side makes sense. I think that's what they're doing, and I think it's a fundamentally unstable pricing scheme.
So I predict that we're going to see corporate workstation users going with xp home and open office. A lot of computers that have been sold with $375 worth of microsoft software on them will now be sold with $94 worth of microsoft software on them.
MS-Office still makes sense for a lot of people. If you run exchange server, and want to use outlook as a groupware client, it makes sense. Excel users who earn a lot are going to get the spreadsheet they know and want, no one's going to tell a $150k/year guy to learn a new spreadsheet. But those types of users don't add up to a monopoly.
If the office monopoly begins to crack, it will be a really big deal. It will be a decline in a core microsoft business, and will suggest that perhaps the best days are behind them. And it will be the result of an open source project.
Windows to linux is a very wrenching change, in a million little ways. But MS-Office to Open Office is a lot more doable.
I think that's where MS's empire will first start to crack.
I've been reading the Linux predictions for every new year, and every year, time passes and nothing revolutionary happens.
I remember in 1998 when Linux was supposed to "surpass Windows on the desktop." I've been hearing it every year since, as well as Linux being "desktop-ready."
Please, hurry up! I really, really want to use it and not be forced to go to OS X for a UNIX desktop. But I have a feeling we'll be stuck in GNOME/KDE world for another 5-10 years.
"Sufferin' succotash."
Here's a thread at Ars' OpenForum giving their predictions. whiprush's initial post is very insightful.
That statement is pure and total bs. Linux got where is it *because* it proved itself by the same measures as other IT components (and it surpassed them as well). If it hadn't, it would never have gotten as big as it is...
If you want to run XP, install XP Gestures is a geek fad, most 'normal' users have never heard of this and never will. The simple dropdown menus are well known and appreciated, tried and true. We don't need to re-invent the wheel, but to improve on it.
The 'radial pie menu' would be the most logical next step...if it can be kept from being intrusive and ugly. I really don't find it difficult to move the mouse around, no matter how far across the desktop I must go...
"Some things have to be believed to be seen." - Ralph Hodgson
What the hell does that have to do with "Linux Predictions for 2004"?
Ummmmm, I predict that parent post will appear in 2004, just as it did in 2003?
Just working from memory mind you and even predictions about the past are risky, but it's kind of a hard post to forget given the specific detail it contains.
KFG
Our brainy heroine and penguin loving paralegal babe, PJ at Groklaw, posted an article covering some New Year's trend spotting. Some of the goodies:
/., but I like the compilation of them as a converging threat to Microsoft's paradigms that may cause significant rethinking in 2004.
:-)
1. Invester's Business Daily makes up its Top 10 Tech Stories of the year without mentioning Microsoft in any context.
2. A speculation comes from Chris Gulker in an IT Managers Journal article that Microsoft will introduce an MSLinux when Longhorn turns out to be unsellable. (Good thing or bad thing? I think good, if it happened.)
3. The example of Smart Displays, where per-user licensing inhibits even Microsoft's innovation, as cited in a Register article:
"The final nail in its coffin was Microsoft's absurd decision to kow-tow to the tin god of its licensing agreements. If you took your smart display downstairs, nobody in the den with the computer could use it. Single user licence, repeated Microsoft marketing droids. 'We can't compromise our standard licensing policy."
4. From the counter example of what can be, in the MagicBike project of the Parsons School of Design, PJ muses: "The idea is, when everyone gets to play, innovation is the result. Innovation doesn't come from money or walled-in projects, although money can help implement ideas. Innovation comes from people, and as George Bernard Shaw once pointed out, talent can show up simply anywhere, where you least expect it. The lower the barrier to entry, the more likely you are to get wonderful ideas. It's one reason I keep it possible to leave anonymous comments on Groklaw, despite the down side to that."
5. Vince Cerf's vision of the ubiquitous net is cited, reaching even to other planets.
PJ concludes: "Yes, [Microsoft] must adapt in order to be part of the future. I think it's a given that no one wants a wireless product that can only legally connect to one PC predetermined during setup. Not after somebody sent the mayor an email from a bike in Union Square station in NYC. Or even read about it. Once you have the concept and you see what is possible, you know what you know, and Brand X doesn't work for you after that. Like the song says, there's nothing like the real thing."
I know most of these points have been previously featured on
Besides, I think I have a crush on PJ...
Any technology distinguishable from magic is insufficiently advanced. - Geek's corollary to Clarke's law
What Linux standards ? How many differnet "Linux" distros are compliant with these standards ? what are they ?
Furthermore, how would it benefit Microsoft to tout that "longhorn is compliant with xx". Microsoft already has source level compat with much free software via the Services For Unix Interix SDK. Windows can be an NFS client or server with SFU. CIFS interop between linux and windows could be better I suppose, but my feeling is that samba needs to move upwards, and microsoft has little incentive to move downward to acheive this.
I guess i'd just be curious to know where this statement came from. It sounds mostly like a "wouldn't that be nice" without a lot of thought behind it.. like an emotional victory rather than something of technical significance..
My opinions are my own, and do not necessarily represent those of my employer.
I figure I'll throw in my two cents:
1.) Package format becomes a hot topic. Discussion regarding a standard takes center stage. Work begins on a standard package format, a stable version is expected in 2005. Adoption of linux on the desktop continues to be slow.
2.) Resolution and refresh rate changing on the fly (ala Windows since 9x) will finally appear in desktop distros.
3.) NTFS read/write support will be sorted out using the NTFS driver from windows. Microsoft will not issue a patch that breaks compatibility, suprising more zealous MS haters.
4.) Gnome or perhaps Mozilla will have servers compromised. The compromise will be found quickly and dealt with quickly as well. Many will use the compromise to point out that UserLinux should have indeed used KDE.
5.) Not linux, but still a prediction. Apple releases some much hyped product with relatively large mind-share. Product is recalled due to design/manufacturing error. Apple faithful blame someone else; Apple stock and market share dip.
6.) Microsoft releases DirectX 10. Doom 3 is the only major linux-native game released in 2004.
7.) Adobe or Autodesk release linux versions of Photoshop or AutoCAD respectively. The released program is quite successful. Many businesses stop using Wine or switch over to linux for their workstations.
Open Source, on the other hand, works according to another economic model, one which is not limited by profit-loss ratios and ROI. If you have people interested in it, you can create an internationalized version of a package for any audience. Now, there are still complicated technological issues (such as some of the really complex scripting systems in many of the smaller markets like SE Asia), but once we get past some of the difficult hurdles of creating truly flexible font and glyph servers and text rendering systems, we will see Linux and FOSS expanding into places where MS cannot hope to go. True, these won't bring in gobs of cash for Linux developers and ISV's, but I think we will see steady progress made. We will soon see Linux as the foundation for technological, and ultimately economic freedom for the majority of the world's governments and citizens.
Your Servant, B. Baggins
KDE 3.2
Xfree86 4.4
Xouvert
A bit pessimistic, I think. KDE 3.3 is scheduled for december 2004 if released on time, and talks are going for making it only 3.2 and half, to give more time for KDE 4.0. That would mean summer 2004. Xfree86 4.4 is already in RC2, with the core-team out the way 4.5 should be out by the end of 2004 too. Also Xserver at freedesktop.org is way more interessting than Xourvert, and should make a lot of splash in 2004.
The last time I used Slackware, package management amounted to little more than tar -zxvf. Has that changed?
No, do you need it to?
One of the nicest aspects of RPM (and perhaps its least used feature) is the way you can manage installed packages. RPM lets you track what is installed. It allows you to verify what is intalled (check for permissons, md5sums, size, etc). It keeps track of versions. It is a breeze to uninstall packages.
Slackware allows you to do all of those things, and do it with standard shell tools.
The thing I like about Slackware's package management is that I don't have to learn a new tool to do everything.
that linux will get some steam, but will face new legal concerns, IBM and sun will be sued by microsoft for imitating their gui, then gnome and kde will be sued in the whole ordeal, apple turns around and sues microsoft ripping off NexT
might not be that far.
but I predict this year will be the year of tech lawsuits as a new major player gets involved in the market.
I wouldnt be surprised if the GPL got overturned and claimed invalid by a well paid judge.
the shit with SCO last year might prove to be the stone in the pond that started the ripple effect.
I know that many business operate with that sort of attitude. I have friends that say "My department has a budget for licensing and related costs - if I don't spend it I will loose it. Besides, if I have any problems, I can just call tech support and they can bail me out. Seems like a pretty good investment to me."
This may all be true but I can only speak from my personal experience. We have a large mixed-platform network (~10,000 nodes running Windows98-XP, MacOS 7.6-Panther, RedHat 6.2-9.0) that is distributed accross about 35 locations. Being a non-profit organization we are constantly hit with ever-tighting budgets and a tiny IT staff.
Over the past several years I have run into many situations where licensing has "cost" way more that it should. Rather than go into an overly-broad generalization, let me give an example that occured last week: I was tasked with upgrading our backup system which had reached capacity. After doing some research on the larger players (NetVault, ArcServe, etc.) I chose an "enterprise-level" backup solution. My particular choice was heavily weighted because our backup system is critical and after the nightmare we had with Welchia, et al, I wanted our backup server to run on Linux and be able to back up to a multi-terrabyte RAID array we'd purchased. I also wanted a "universal management client" that would run on any of our client platforms. Anyway, I won't go into whom I chose, but the problems I've had since then have almost ALL been related to licensing.
First off, the "core" program is reasonably priced. But every client that you are going to back up has to have client software loaded (requiring a reboot unless it's linux - so now I have to wait until after hours to install) Each client comes with it own license that's tied to that machine. OK, no problem. I'm just doing initial testing and each client has a 30-day evaluation license (except it doesn't take for some reason - call tech support - they can only give me a 7-day eval license key - WTF? - OK, I'll use that.)
I run a test backup and it misses about a thousand files - Oops, silly me me "Open Files" require a separate plugin (that costs as much by itself as the client plugin.) So I install that plugin and re-run the job. It still misses hundreds of files. I investigate. All the files were saved to the server by older mac clients and had "invalid characters" in their name (like forward slashes, etc.) Strange, all our mac and PC clients can access these files - why can't the backup software? (Call tech support - I need the UNICODE plugin for windows 2000 Server. It costs as much as the original client software. This particular server is NT 4 - Sorry, no plugin for you! You must upgrade to Server 2K ($$$) and then buy the plugin license. WTF!!!)
I don't WANT to upgrade that server. It's just a file and print server. It does everything I want it to do the way it is. Besides, I'll have to buy server 2003 for it - How will M$ licensing affect me then? What other hidden costs will I have to prepare for in my uncertain budget?
OK, I can use our existing backup solution to back that server up until we get it upgraded or migrated. By now my 7-day keys have expired so I can't do any more test backups on my other servers (MsSQL: separate plugin. MysQL: separate plugin. Filenames beginning in 'T': I'm sure I'll need to buy$$$ a separate plugin.) So I figure I'll just register the backup server and the five-or-so clients I want to do the longer term tests on. I head over to the registration page and fill in the information and am about to hit submit when I notice some small print between the last field and the submit button. I actually read it :-) it says, in part:
"terrorism" and "pedophilia" are the root passwords to the Constitution