Slashdot Mirror

We have 140 Linux based X-Terminals up and running, and this is year 3 of the project.

It is possible :) Props to LTSP for making it almost easy :).

Alright, so Microsoft is trying to strangle out interoperability. And this is new because?.. ;)

Just my two cents, but personally I think that we're barking up the wrong tree here. Even if this new .NET centerpiece causes serious interoperability problems, one has to take note of Microsoft's other decisions; the fact that its currently shooting itself in the foot with licensing for instance. All good sysadmins (yes and MCSEs too) know that you shouldnt jump on bandwagons, .NET wont be seriously pervasive until years after its launch, so maybe instead of going on about how Microsoft is leveraging its monopoly blah blah blah we ought to look at Linux's strengths. Like terminal services for instance. Microsoft didn't invent that as they might have you believe - get a few old P75's, network them together, put an application server and you have a centrally adminstrateable server with almost zero-maintainance cost. Want to do that with Win2K? sure, after you fork out thousands of $$$ for something that isnt much better, if at all. KDE's fast becoming a serious desktop platform, together these two capabilities could really make a decent corporate solution. And if you think this is a pipe dream it seems to be good enough for a whole city to use. (yes I know they still use Citrix machines, but not as much as the KDE boxen hmm?)

And that's just one possible application. Linux seems to be doing everything except making coffee these days (hell, even that) and more and more people seem to be using Linux these days in both client and server space, so screw Microsoft. People dont use NFS or Apache because its compatible with MS proprietary stuff. Learn from history too; OS/2 tried to be an advanced platform as well as being Windows compatible. Don't see OS/2 around much these days, do you?

I'm sure some of you will disagree with me but in my humble opinion the mono effort isnt a great idea. Why not implement an open source .NET? thats what we should have been doing from the start, instead of following behind. That or jump over .NET and do the Next Big Thing, whatever that may be. Or stuff compatibility and instead focus on making a better open UNIX platform, people arent falling over themselves to shell out for MS' latest technology you know. I'm losing my point anyways, so enough karma whoring for one day, but never theless, I suppose this is worth thinking about

I think that .NET is an interesting idea. But I'm confused why people consider this a revolutionary idea - ok, so MS has all your apps on the central server. I think I must have missed something, because most of the primary software (OS, basic functions, etc) are run locally.

Whereas with some of the more interesting projects - that have come to light lately - we are making *significant* headway into the 'NC' idea that was oh-so-trendy about four and a half years ago. The idea of HD-less boxes has become available to the masses. The Linux Terminal Server Project has managed to compile an extensive resource for creating NC's that are feasible for businesses (and significantly more efficient to administer than either a set of Linux boxes or Windows boxes.) K12LTSP is also on the way for schools.

Will either of those projects reach wide-adoption? Not bloody likely. But maybe I'm missing some grand-paradigm of ASP's that can't be fulfilled by a well-designed client/server relationship. I don't think I would ever invest in .NET when the blatantly obvious alternatives exists - especially when I can get away for $350 per client box.

Rather than a 5 Watt pot (which is going to be a bit expensive) use a zener diode. I've had very good results with zeners in the range of 7-8 volts. You can pick up 10 of them for about 3 bucks. I've modified several of the LTSP terminals we use where I work, and they do the job nicely. Scott

Wow, must have been hard to write the article when you have stuff like this, In Issue 27 of the Linux Gazette even.

Other sources:
Diskless HOWTO
XDMCP HOWTO
XDM-Xterm Mini-HOWTO
Linux Terminal Server Project

I'm suprised it made it past the Linux Gazette Editors actually, considering it was in Issue 27! You've got one year of HS left, but I suggest you try harder in college or someone will call you on plagiarism. I admit however that I didn't actually read the artice, as I'm not likely to find any new information.

So, what about the Linux Terminal Server Project. Isn't that all about X-terminals. What's to important about this "News". The benefits are known.

This page by Jason Spisak has some information about building a thin client with no moving parts. It includes an address to TK Power which manufactures fanless PSUs. Depending on your systems power needs, a fanless PSU just might be what you need. -K

well, for server tasks a little spare resources are nice ... but as far as clients are concerned: have a look at The linux terminal server project. NIC, boot-rom, some paint and you have a nice terminal. cu grissu

We love old 486/Pentium boxes here! Our firewall runs on freesco and thanks to the LTSP project, we have equipped everyone's desktop with a low-cost X terminal. There is a write-up of what we have done for anyone who cares. The beauty is that we have incredibly low cost-of-ownership, don't care if anyone breaks in and steals the stuff and it is totally silent in operation. The biggest complaint in the office now is the noise of the damn clock ticking. It has been a wonderful experience, they don't break down, you can boot one up from anybody's desk and get your own desktop ... send me as many as you have got!!

I think the whole issue of Linux acceptance outside of the server realm comes down to this: there is TOO much choice in how to set up Linux.

The Linux Terminal Server Project provides one solution to this. As far as IT issues go, it should provide a "win-win" -- you get to use standard, cheap PCs as a "thin client" AND you get the centralized control over everyone's configuration that IS loves so much (and that does help lower TCO).

For a look at how this works in the "real world," read Major Law Firm Installation of Linux. This provided a standardized KDE desktop for the administrative staff. Since the customer was a long-time WordPerfect user, the staff did not require application retraining, and only required minimal training on the Win* to KDE conversion.

This also had the nice effect of changing an IT manager's nightmare into an IT manager's dream. The law firm ended up with a single point of control for all their desktops, which they could then even oursouce many operations back to Unique Systems, the company which did the rollout in the first place.

What IT manager wouldn't like to be able to say to their boss "Look, for a small consulting charge and minimal retraining, I managed to cut our license fees AND support costs, preserve our legacy data and applications, outsource our administrative overhead, and I did this without purchasing any new hardware"?

(And no, I don't work for Unique Systems, I'm just familiar with them from them from their good work with the Toledo Area Linux Users Group, and from considering just this setup for a former employer.)

See EBay, or just stroll in to Target and plunk down $299. It's a fine X terminal running Jailbait, or check out the Linux Terminal Server Project if you don't want to put a disk on it but do want more than what JailBait provides. Or check out the bboard ; ; at http://www.linux-hacker.net for details on everything necessary to turn an I-Opener into a standard Linux machine.

http://www.ltsp.org?

Next, the xtermkit I used in our project - it's written by Jacques Gelinas of linuxconf fame.

And lastly, check out the Linux Terminal Server Project. Another form of basically the same concept.

Check them out - I fully believe X terminals are a wonderful way to reduce costs - hardware, software, and management - when compared to more traditional desktop computer systems.

You will want to check out the linux terminal server project and maybe also this site.
I suggest not having hard drives. Go with a bootrom or an etherboot-enabled floppy disk. And definately go for full-duplex ethernet. 10 or 100 mbits, depending on what kind of hardware you can get.

OK, I'm sorry if I misinterpret your statement.. But it seems to me like you're saying that a PC is a thing of power. You have your fixed disks that you can modify the data on, as opposed to this little Thin Machine that gets all it's data read-only from a main system somewhere.

Well, you're right. Mostly. A PC is a good thing. But let's face it, a modern PC has more power and more hassles then a lot of situations need. Say you need to set up a lot of cheap kiosks? Read the Long Story secion at LTSP, it's a very good example of why those 'powerful PCs' just don't suit the purpose sometimes.

Maybe you ARE the sysadmin and you don't want all your lusers to have huge amounts of power on their systems. Say at a school. Most High School students (not the good geeks, the kiddies) will take your systems and do their damndnest to destroy them. Don't give them that option, use thin clients. Have control, have every system cheap and easy to diagnose.

Or maybe you just want to get on the 'net? That's good. You just either need to put up with the fairly easy-to-use-but-unstable Win9x with it's useless Hardware Wizards and Paperclips, or learn Linux (no easy feat). You know something, neither of those options suit some people. Embedded systems are simple. They can't be fscked with by the user that greatly.

And as for your very weak argument on a price point, look at the i-opener. Sure, it's being sold at a LOSS, but it's cheap as dirt. x86 may not be the ideal embedded system, but it works. And other archs aren't as obscure as you make them out to be. That's what this article is all about, the increased availability of other archs.

..an easy way to make an embedded-ish system is with Boot ROM's on your ethernet card! NFS your root FS, get your kernel over the network.

Read up on the Linux Terminal Server Project. It's good stuff.

Also check out the The Award Winning Story from the Linux Terminal Server Project.

3. Target other markets: schools, colleges and universities could use cheap machines with standardized, open-source OS installed. Target large corporations, who need a computer on every desk, and sell them these machines. With Linux or one of the BSDs, you can overturn the Microsoft monopoly.

Why a hard drive? All you need is a kernel in flash with USB ethernet drivers (2.3.99pre) and the Linux Terminal Server Project server. Mount your filesystems
from the server.

Anomalous: inconsistent with or deviating from what is usual, normal, or expected

Here is another I-Opener hack. You can now boot linux from the flash memory instead of disk. The problem is you still need a disk and cable to make the initial transfer. Check it out here.

I was trying to do something like this at a community center our LUG was helping setup. Right now they are using Windows95 as a microkernel for StarOffice, but as you all know Windows is anything but "micro".

What I wanted to do was migrate them to was very simple local installations of Linux, X and SO only (maybe using remote boot images from the Linux Terminal Server Project. It would have been great, I could have them boot and load SO from a local HDD but have their home directories and mail on the server. SO can do mail, calendar, web browsing, Java, as well as word processing, spreadsheet, presentations, database, etc. so the local admin would only have one program and file tree to maintain.

Unfortunately this won't happen now because they hired a Windows guy to admin and teach, who doesn't know the first thing about Linux. He got frustrated the first day (probably a permissions thing on the Samba shares, maybe he wasn't logging in as himself, who knows?) so he fdisked and formatted the server, without even asking us for help. He is using Win95 File and Print sharing instead of a Samba server, what a maroon.

Maybe good ol' Larry could take a look at Linux Terminal Server Project here.
It has not been done by what I'd call Linux experts (ahem), but sure it is interesting.

Now, if I only could find an IBM RIPL implementation that works on Token Rings cards... 8-)

Bye, Rob!