Slashdot Log In
IBM's Linux Upgrade Roadmap
Posted by
michael
on Sun Mar 28, 2004 12:42 PM
from the cheat-sheet dept.
from the cheat-sheet dept.
petrus4 writes "IBM have put together a nine-part series on upgrading from various incarnations of Windows (NT in particular) to Linux. Although it's mainly aimed at corporate customers, it's a good read, and could help the Linux advocacy effort in general."
This discussion has been archived.
No new comments can be posted.
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
Full
Abbreviated
Hidden
Loading... please wait.
IBM (Score:5, Funny)
Nice to be backed by IBM ... (Score:5, Insightful)
I get that same warm, tingly feeling inside as did the members of Team OS/2 in the old days.
Parent
Re:Nice to be backed by IBM ... (Score:5, Insightful)
The optimist in me is hoping that IBM will stick to its guns this time.
There is more support now, and if you remember, it was difficult to get systems with OS/2 preloaded on them.
Linux has more marketshare, i think, and definetly more mindshare.
I think Linux will clear the hurdle....
Parent
Re:Nice to be backed by IBM ... (Score:5, Interesting)
As was I -- scoffing, I tried it for a competitive analysis that marketing wanted and was hooked within 2 months. It was damn nice for the time, and some features in the WPS would be nice in KDE and/or Gnome if not at the window manager and -- better -- file system level.
That's the beauty of it. It won't matter in the long run if they do or don't!
Nearly impossible. Dell, Gateway, Compaq, and even IBM never preloaded OS/2. Now, we're getting preloads...and even companies like HP and IBM yelling that they are the biggest Linux supporter. Dell brags too...though I'd like them all to shut up and get the goods out there.
Linux isn't being laughed at. OS/2 was only taken seriously at the begining, and quickly became the ugly step child (though technically it was quite nice though prone to crash/lock the UI).
I'm starting to see job listings specifically asking for Linux experience crop up in various places. My OS/2 experience never seemed to be important to anyone.
Parent
Re:Nice to be backed by IBM ... (Score:5, Informative)
Untrue. It's still fairly easy for a program using XFree86 to freeze or crash the entire Xserver, killing any other process using it.
It is especially easy for one program to accidently starve all others of input- this will happen, for example, if a Motif program freezes while a menu is pulled down. (In a case like that, a user with an alternative means of access can kill the offending process remotely- but only experts can do that, so this case must still be counted as a severe failure of interprocess protection)
The only best way to fix this problem would require major changes to the X11 protocol- probably big enough to deserve a major version increase up to R7. It should be possible for applications to survive the shutdown/crash of the display server they are using, and attach to another one later. (Protocols like VNC and RDE allow that to a certain extent, as do some TTY consoles; X11 should too)
Parent
Re:Nice to be backed by IBM ... (Score:5, Interesting)
They stepped on Bill's weener years ago so M$ is NOT an option.
IBM has put ALL of it's eggs in the Linux basket. It's sink or swim and the ONLY life preserver available to IBM is Linux.
Parent
Re:Nice to be backed by IBM ... (Score:5, Informative)
Parent
Re:Nice to be backed by IBM ... (Score:5, Interesting)
Parent
Which guns? (Score:4, Insightful)
Another reason OS/2 was doomed from the start: people don't like to buy technology from their competitors. That's why AT&T finally had to spin off its manufacturing arm, so it could sell stuff to competing phone companies. I don't know how hard IBM tried to get Compaq or Dell to bundle OS/2, but it would have been a hard sell.
As for Linux, IBM hasn't yet manned all the guns there either. They're selling it strictly as a server OS. You hear noises about them moving to Linux as a standard desktop, but so far these are just noises -- every IBM laptop, desktop, and workstation still comes with Windows pre-loaded!
Parent
Forget the cold cash (Score:5, Interesting)
But before you get too enthusiastic, remember that they're treating Linux strictly as a server OS. Go to their web site and try to find a single desktop, workstation, or laptop that does not come bundled with Windows. You don't even have the option of buying the system witout an OS!
Parent
Re:Forget the cold cash (Score:4, Insightful)
This is where I agree. Linux developers should be concentrating on the server market. This is where they have the most going for it now. Fast, Stable, Secure should be the motto. Desktops will come soon enough.
Parent
Re:IBM (Score:5, Insightful)
Parent
How to buy a IBM Thinkpad with Linux? (Score:5, Interesting)
Re:How to buy a IBM Thinkpad with Linux? (Score:5, Insightful)
But their opinion, and the opinion of most non-slashdoters, is that Linux isn't ready for the desktop now for home and many coporate users.
This isn't a flamebait. It's just that the article isn't supposed to answer these questions. XP professional IS what they recommend and for a good reason. Support for wlan isn't IBMs problem. Servers don't need to have a wireless network connection. If you want support for Linux Hardware from IBM, go here [ibm.com].
Parent
Re:How to buy a IBM Thinkpad with Linux? (Score:4, Informative)
Parent
Slightly disingenuous (Score:5, Insightful)
While I look forward to the day a Linux distro can upgrade an NT system, carrying forward system settings, user passwords, domain logons and applications carried across into WINE, this isn't happening anytime soon.
I'd be surprised if it doesn't happen eventually, though.
Re:Slightly disingenuous (Score:5, Insightful)
well all this requires is a program that saves settings from NT and restores them into linux, without neccessarily upgrading your NT partition to linux. although this would be more useful to the corporate world as home users probably wouldn't need this i would imagine.
Parent
Re:Slightly disingenuous (Score:5, Insightful)
Parent
Re:Slightly disingenuous (Score:5, Interesting)
At the moment it looks like a pretty custom job, but I can definitely see a generic tool being in the works.
-Mark
Parent
That's a training guide, not an upgrade roadmap (Score:5, Interesting)
Suse vs. Blue Linux (Score:5, Interesting)
Re:Suse vs. Blue Linux (Score:5, Interesting)
So if IBM is now going to be using Suse, does this mean that the Blue Linux rumor is bogus?
Speaking as a Linux user inside IBM, I always took the view that the Blue Linux rumour was bogus, or at least misleading. There is linux software flowing around the internals of IBM - plenty of it. And we do have various packages nicely wrapped up in RPMs that aren't available outside IBM (Lotus Notes 6.51 running on a standard WINE base springs immediately to mind). However, an internal distribution is a far cry from launching an external distro. That's not to say it will never happen. Just it doesn't look likely imminently.
... and yes, my sig DOES apply here. I'm not an IBM spokesman reading some approved script.
Cheers,
Toby Haynes
Parent
Upgrade? (Score:5, Informative)
We're assuming you already have Linux installed.
While the article does point you to linux.org for choosing a distro and whatnot, any good guide to switching to linux should at least cover the basic installation methods and what you'll need to think about before installing. Since distros have different installers, you don't want to get too indepth or focus on any one installer, but it should at least cover ideas that would be universal or at least common to getting ready to install linux.
Superb (Score:5, Insightful)
/grin (Score:5, Interesting)
["If you have been using Windows for a long time, you are accustomed to rebooting the system for many reasons, from software installation to correcting problems with a service. This is a habit you will need to change to start thinking in Linux. Linux tends to be rather Newtonian in nature. Once set in motion, it will tend to stay in motion until it is acted upon by an outside force, such as a hardware failure."]
Re:/grin (Score:5, Insightful)
The Linux community needs to stop hiding behind "we don't have to reboot" - it's just not as compelling an argument as it was 2 years ago. And with clustered servers becoming the norm, (MS AppCenter, etc.), reboots are hardly even noticed by the end-users.
Parent
Re:/grin (Score:5, Insightful)
What Linux version are you running? If it's 241 days old, then it is probably missing quite a few kernel security fixes.
Parent
IBM Solution to Linux in 3 Easy Steps (Score:5, Funny)
2. Install Windows OS and any related applications
3. After the IBM drive crashes and destroys your data (6 to 8 mos), you can install Linux without worrying about prior data.
console (Score:4, Insightful)
In fact, I just finished installing a wireless card in my linux box. Comprared, to windows, where I pop in a cd and hit install, under linux I had to:
1: make & make install the software
2: install some necessary wireless libraries
3: manually configure the wireless card's config file
4: set the kernel to intialize ath0 at startup
now to a techie following a recipie, this is a piece of cake. However, it is quite beyond the capabilities of your avg windows user.
Re:console (Score:5, Insightful)
OK, I use Slackware at home, so I have all the fun you're talking about, but when I install systems for people, I install Mandrake, and I have never had easier hardware installation (and yes, I have delt with XP systems).
A new printer? plug it in, turn it on, control center, click on printers, wait 30 secs, print a test page just to be sure....
Netcards are just as easy, turn the machine off (sniff, so long uptime, I new you well
I'm afraid your either misinformed, or trolling, because the hardware question has been well and truly delt with.
Linux still has some technical hurdles to overcome before being a Joe Sixpack desktop machine (some cups quirks come to mind, where if you print to a turned off printer, you can only print to it again, if you reset it in the cups web interface, and I know there are other problems), but this is just 3 year old FUD.
Parent
faulty windows/printer vs linux/net analogy? (Score:4, Interesting)
Q: is the quoted idea historically accurate, given the development of unix, which begat linux?
nt domains upgrade is easy. (Score:5, Interesting)
the migration tools for nt 4 style domains would take about 2 to 3 weeks to do: most of the work has already been done, it's a matter of documenting it, checking it and making it easier to use.
the open source migration tools for nt 5 (aka w2k) style domains would take a bit longer: a few months, at most, though, as various efforts (e.g. heimdal) are already underway.
the open source migration tools for exchange, now _that's_ a challenge, requiring about twelve to eighteen man-months of work to get somewhere.
i know someone who has done most of the work already, in his spare time: it's proprietary but if an open source exchange project was to seriously take off, i know he'd consider releasing some of his code to 1) help out 2) make sure _his_ copyright notices are at the top of the files, because in open source just as in the proprietary world, the _first_ person to release is the one that tends to take off, not the best.
ironically, just ONE company with more than one hundred employees that will be looking to pay microsoft's next set of exorbitant upgrade-because-everyone-else-has-and-oh-look-eve
Mmmm. Nine-Part-Series (Score:4, Funny)
1. Insert disk. Answer questions.
2. The usual ???
3. Profit.
Upgrading from WinNT to Linux:
1. Delegate to line manager to hire transition team.
2. Study IBM's 9-part series on upgrading systems.
3. Days into weeks into months of unusual ???
4. Hold post mortem on cost overruns and continuing bugs.
5. Fire transition team and manager.
6. Hire consulting firm to manage systems.
7. Look for ways to cut costs elsewhere to cover P&L ass.
8. Hope nobody asks ???
9. Loss.
Re:Mmmm. Nine-Part-Series (Score:5, Insightful)
It's called long term planning. Sure right this instant it may cost more to move to Linux from WinNT. However, what about when license renewal day comes around? What happens when WinNT is no longer supported [e.g. no patches for the day-to-day exploits?] etc, etc, etc...
In the long run the average linux distro [say Gentoo] will cost a hell of a lot less.
And hey, if it requires the users to learn a bit about computers is that really such a bad thing? I mean for the most part people can just use KDE and be happy for it. For other things they can learn the fun way, google for it.
Tom
Parent
Re:Mmmm. Nine-Part-Series (Score:5, Informative)
To start with, because the interface is so similar, plus "Hey it's just Windows!" comments from superiors means that few are taking the time to learn the intricate differences. (such as permissions and account handling)
Then you get the broken programs.
Then you get the boss who has [what he calls] critical data in an older version of Access that you must now move to [new] Access, which seems to be impossible in certain (read many) cases.
Or, I can implement an IBM-driven Linux-based solution that would force superiors to treat it differently, plus I would have more control over whether or not to continue on the upgrade path to future versions. Microsoft doesn't give me that. To stay secure, even using their loose definition, means continually upgrading, breaking software, data, and perceptions all the while sending them more money.
Oh, how I wanted to get out of that cycle at my last job. Now, I might be stuck implementing it, but at least I'm not responsible for the mess my superiors make trying to fall in line behind Microsoft.
Kudos to IBM for making it reasonably easy to know what's in store for those trying to get away from Microsoft.
Parent
Ultimate Killer App (Score:4, Interesting)
0) Install spare HD and set BIOS to boot from CD, restart.
1) Knoppix-based CD boots the server
2) VMWare installation on CD boots the Windows OS from the HD on top of Linux.
3) Various scripts portscan the VMWare-running server and scans the filesystem for info, creates a Linux installation on the empty disk and copies all services and shared files to this new installation. Creates Samba server to host login/password info if needed (PDC). Copies Exchange server, IIS, DNS etc. etc. Shutdown when finished.
4) Swap the old intact primary HD with the brand new disk and restart, booting the new Linux clone. Test and apply any manual changes if needed.
5) Sell these scripts as Linux Migration Kit.
6) Get sued.
Misleading title (Score:4, Informative)
We need more of this! (Score:5, Interesting)
Funny how times change. (Score:4, Interesting)
All the 'little' people wanted them to be taken down to size, releasing the hardware to the people.. 'freedom'...
Now we root for them as they may just save the OSS movement from the giant beast...
Re:Funny how times change. (Score:4, Insightful)
So nothing has really changed; monopolies are the enemies of everyone who is involved in a market. Nowadays of course, we have rather better tools with which to fight monopolists. Balanced against that, unfortunately, is an unwillingness for governments to fight monopolies effectively.
Parent
Re:roadmap?? (Score:5, Interesting)
The list IBM covers is quite handy for training others who think Windows is the beggining and end of what computers can do.
(That said, I would be careful using Webmin -- Step 3 -- as it can cause problems, though as an introduction when a more experienced admin is around it is OK.)
Is this useful for you and me directly? No. Can these texts help us by making conversations with the Windows-obsessed but willing less frustrating? Yes.
Here's the list from the link;
The first step to success in Linux is learning to think in Linux. Take what you already know and redirect it to doing things the Linux way.
Step 2. Console crash course Linux provides great power and flexibility through the console. If it has been a while since you've spent much time at the command prompt, take a little time to reacquaint yourself with this environment by reviewing common commands you'll use all the time.
Step 3. Introduction to Webmin
While it is important to know the nuts and bolts of administration, it is often more convenient to have a tool. Also, a higher-level application makes complex configurations easier to handle. Webmin provides point-and-click configuration for beginning and experienced administrators.
Step 4. User administration If a system has no users, is it really a system? Learn about the Linux approach to users.
Step 5. Linux logging Linux makes extensive use of logging. Nothing is hidden from you. Becoming comfortable and familiar with logs will allow you to monitor the health of your system and track activities.
Step 6. Working with file systems File systems are at the heart of every server. Linux provides a lot of flexibility in its file systems.
Step 7. Networking
Working unconnected is unthinkable in today's world. Linux on the network unleashes its full potential. However, Linux networking looks very different on its face. You'll need to learn some new terminologies and new tools.
Step 8. Backup and recovery
The first line of defense against disaster is a backup of the data. Linux provides different options, some of which are very simple to work with.
Step 9. Installing software
Linux can use prepackaged binary files, or you can compile programs directly from source code. The tools for installing Linux programs are very useful and provide functionality you might not expect.
Parent
Re:roadmap?? (Score:5, Funny)
Yeah! Roadmaps are for sheep! Real Men pretend to know where they're going!
Parent
Re:Good to See (Score:5, Funny)
Change businessmodel from:
1: Do stuff.
2: sell it.
3: Profit!
To:
1: Do stuff.
2: Give it away for free.
3: Hmmm.....
Parent
Re:Good to See (Score:4, Informative)
Parent
wrong business model (Score:5, Interesting)
1. Do stuff
2. Report plenty of bugs, RFEs to MS for free
3. Pay annual licensing fees to MS
3a. Hope that Microsoft won't screw you by making changes to their s/w that help their bottom line but hurt you
4. Hope enough money is left over from your core business
With OSS s/w, this becomes
1. Do stuff
2. Report plenty of bugs, RFEs to OSS project (occasionally fix/implement one) for free
3. Make profit from your core business
In both cases, you do free work for other people, but with OSS, all the free work is aggregated and you don't pay for it over and over again. With MS, you end up paying for the same piece of software and for the volunteer work of others and yourself not just once, but over and over again. Furthermore, with MS and other commercial s/w vendors, you constantly run the risk that they will screw you by discontinuing or changing products you depend on, and you have no recourse.
The business case for OSS is easy to make: OSS greatly reduces risks and cost of ownership. OSS isn't without any costs, but it is cheaper on balance.
Note that OSS is a business model and money saver for the actual end users, comapnies whose business is not the creation of the OSS itself, but something else. Founding a s/w company that creates OSS and makes money from it is, as you yourself observe, a long shot and only works rarely. And that's OK.
Parent
Re:wrong business model (Score:4, Insightful)
1. Do stuff.
2. Find problems and bugs.
3. Hope somebody out there at random fixes it.
4. Wait some more.
5. Hopefully still make profit.
The notion of treating your business computers like 'information appliances,' meanwhile, has vaporized. And that's what a lot of businesses are after. Companies don't hire mechanical engineers to build them special-purpose cogs for the copying machine that will make it produce copies 20% faster. They won't hire programmers, either.
Commercial vendors are in a drive toward standaridzation, and working to turn computer software, and the support needed to administer it, into a commodity. The notion of returning to the 1980's method of hiring 'consultants' to engage in special code tweaks on their equipment is antiquated and it's exactly what businesses do NOT want any longer.
Now, if IBM can hide all that activity beneath a 'shroud' called IBM, and certify their team of people to engage in said support activities, they'll get somewhere.
Gonna work as a drone for IBM sometime in the near future? You're not gonna get the contract to work on IBM deployments as an independent contractor.
Parent
Re:wrong business model (Score:4, Insightful)
> standaridzation, and working to turn computer
> software, and the support needed to administer
> it, into a commodity. The notion of returning to > the 1980's method of hiring 'consultants' to
> engage in special code tweaks on their equipment
> is antiquated
BWAHAHAHAHAHAHAH!!!!
I'm sure the massive amount of money being made by Oracle's services division and IBM's and HP's and... Well, you get the picture.
Consultants are NOT going anywhere anytime soon - especially not to India.
Have a nice day.
Parent
Re:wrong business model (Score:4, Insightful)
1. Do stuff.
2. Find problems and bugs.
3. Hope somebody out there at random fixes it.
4. Wait some more.
5. Hopefully still make profit.
You got points 3. and 4. wrong: if you need it, you don't wait, you fix it yourself and return the fixes.
Commercial vendors are in a drive toward standaridzation, and working to turn computer software, and the support needed to administer it, into a commodity.
Yes, and nothing about what I said contradicts that. If you outsource your support and administration, then the company you outsource it to becomes the participant in the OSS projects. That's actually the most common form of OSS usage, where companies like RedHat, SuSE, etc. get paid for easy-to-install (but still OSS) solutions, but they sponsor projects to fix specific bugs and add specific enhancements that many of their customers want. If anything, OSS works better in that kind of world than something like Microsoft.
The notion of treating your business computers like 'information appliances,' meanwhile, has vaporized.
No matter how you handle the low-level maintenance of your software, and no matter whether you go with commercial or OSS, for many businesses, that is suicide anyway. Business software encodes how businesses run; it basically is the business.
You can treat it like an "information appliance" about as much as you can treat the CEO like an "information appliance".
Parent
Re:Good to See (Score:4, Informative)
Allow me to fill in the blanks:
Change businessmodel from:
1. Do stuff
2. Sell it
3. Profit!
To:
1. Do stuff
2. Give it away for free
3. Sell services.
4. Profit!
Right?
Parent
Re:Wrong. (Score:4, Insightful)
A fully multi-user system would be one where both you and your girlfriend can be using the computer at the same time. Letting several users have sessions open is a step in the right direction, but it's not really multi-user if one of you has to stop what you're doing when the other wants to use the computer.
That's not to say the technology isn't there, though. Terminal Services (and 2003's "remote desktop for administration") is properly multi-user. However, that's a separate feature which you have to buy a separate license for, and then manually turn on. With Linux systems, it's there out-of-the-box.
Parent