Yes, that's what I did, and the same for CyberNerd and Geoff I think. The community and wiki are fantastic resources.
Printable View
Mostly I'm self taught, but TBH I did some courses linux when I was at Uni, which I blagged for free as i was looking after some linux servers for my college.
Computing Service course descriptions - Unix (including Linux) Systems & Use
Normally I'm one of first in the queue but Geoff's post
Is one of most informatative, concise, practical I've seen in this forum :D :D :DQuote:
Ok, I here's a practical post on how to move (most) of your IT over to Linux.
1) Replace your server infrastructure first. Start with easy things like DHCP/DNS, File & Print, etc. You can do this slowly in the background without interrupting your users. Leave a couple of Windows boxes as DCs so you can keep AD, but that's all you need.
2) Move to a Thin Client based client/server system. You will obviously need to use 32bit windows servers for this so the software your users need works. However you are free to use Linux on your client devices. Thinstation is a good option here.
3) Mission almost complete. You have 2 DCs + X Terminal Servers running Windows and that's all, and your users are none the wiser. You may now spend your days playing Quake on your fault free systems waiting for Microsoft to fall and Linux on the Desktop to be realised....
(And way ahead of his others - which are usually concise :p )
regards
Simon
PS I'd echo that the cost of MS licenses isn't a big issue (IMO they are value for money in schools) but you can save £30/computer just by switching to OpenOffice which is the quickest win on the planet :)
Wow,
I never expected an arguement to break out when I asked that question.
Initially all I will be doing is setting up a test environment completly separated from my live networks.
I think to start I will set up two severs. One will run DNS, DHCP and user auth.
The other one will run LTSP and will be a file server.
Then I will have two clients. One will be a thin client and one a fat client.
Not sure what Distro to use though. Has anyone used Karoshi. I had a look at the demo and it seemed a bit flaky and the support doesnt seem all that good.
I had thought of using OpenSuse for the servers, but I've ready that it can be a pain to set up.
For the clients I was thinking of using either OpenSuse or Ubuntu.
We run Facility on another network joined by a Router that only allows http traffic to cross the networks.
I've not used karoshi myself, as I prefer to custom build things exactly how I want em.
I'd say use Ubuntu server 8.04 LTS for servers and Edubuntu for clients.
Personally, I'm on a virtualise-everything campaign at the moment, for which I found CentOS to be best. It gives you virtulisation/clustering/replication/etc options right on the install screen. Ubuntu 8.04 wouldn't boot for some odd reason on our new servers, although I didn't hang around to find out why. We'll be upgrading our Edubuntu thin-client server to 8.04 soon, so we'll see how that goes.
--
David Hicks
They'll be using software. Normal users don't sit down in front of a computer to use an operating system, most users don't even know what an operating system is, they simply want to get a bunch of tasks done. In 10 years time, there's a good chance that most software won't even be client-side OS-based, it'll be server-side OS-based and the user will interact with it via a web browser.
--
David Hicks
Ok so Karoshi is off the list.
Karoshi has a lot of potential, but it falls down on a couple major things for me.
First is the lack of online support.
Second is the fact that you can't change use your own IP ranges or hostnames without breaking other parts of the system. I can't believe anyone would design this type of software with such a flaw.
Other than that its a really good peice of software with so much potential.
Next on the list is Ubuntu Server and clients
after that OpenSuse Server and clients.
AFAIK Karoshi was designed and built up by people like us (e.g not professional software developers) who've done a wonderful job that deserves lots of praise and they were in an enviroment (Direct IP to the internet - not tied to an RBOC dictated IP range) that worked for them.Quote:
I can't believe anyone would design this type of software with such a flaw.
You CAN change the ip addresses (check out my posts and Jo's replies on the Karoshi forum about 12-18mth ago) - but it takes a little bit more effort and all it needs is a pre-configuration tool to setup the IP ranges you need but since the developers don't need it - it was never written and tested (AFAIK - times might have changed since I last looked at it)Quote:
Second is the fact that you can't change use your own IP ranges or hostnames without breaking other parts of the system.
But its a very impressive piece of work IMO :)
regards
Simon
PSIs the forum dead??? [EDIT] just checked out site - no forum! and a few other changes that now explains your post![/EDIT]Quote:
First is the lack of online support.
Who will benefit most a full implemenation of Linux.?Will leaners leaving your courses find it easier to get a job putting open office/Linux on their CV, compared to MS office? I am hoping to work with departments to allow leaners to access multiple OSs/Apps but I am tryng to avoid MACs(LOL)
I have implemented several Linux servers, and open source apps on Windows and would love to do more. But I have to balance what is in the best interest of the leaners as well.
Also the thought of staff retraining is VERY scarey! I am still trying to extinguish the use Publisher! Never mind scrapping whole of MS office. (LOL)
Why would anyone put 'Office' on their CV unless they had a specific qualification in using it?
You would put skills such as 'Word Processing', 'Spreadsheets' etc... Unless you had an MCP qualification. Also, most people leaving school would simply list that they had a) GCSE ICT b) A-Level ICT or c) Key Skills as a qualification, which indicates they have skills in the basics of using an office suite.
So put simply, the kids would benefit most, as the money that is wasted on MS software can be put into useful things like more physical machines, or more staff etc...
And why avoid mac's? Apple's market growth has been quite amazing in the last 4 or so years...