Viability of LTSP Wireless Thin Clients

[oortcloud 0]

I'm currently in the process of developing a set of wireless thin client network solutions using Linux LTSP servers and VIA mini-itx thin clients (Orinoco Silver PCMCIA wireless card). I've looked at the economics of this approach for businesses and home networks and the cost savings over a 3-5 year time period using the Linux thin client approach seems like it should be quite compelling. The performance I'm seeing using an 802.11b thin client against a Ximian Desktop2/Redhat 8 instance on top of VMWare is passable. On dedicated Linux servers the wireless thin client performance is quite solid.

 

My question is: Can we (Linux advocates/businesses) make further in-roads and take some M$ general purpose desktop territory by offering homes and businesses wired and wireless thin client solutions? Will the economics be compelling enough? My argument is the management of multiple M$ PC's over 3-5 years (especially upgrades and migration of local data) makes the thin client network solution a very strong offering. I wrote a white paper (of sorts) describing my view on the subject located on my commerce site. Comments, suggestions on moving forward are welcome.

 

My commerce site is: http://commerce.oortcloudcoding.com

The LTSP site is: http://www.ltsp.org

 

P.S. I used to admin a network with IBM XStations and I loved them. If they broke you threw it away and plunked down another one. I guess I'm kind of partial to this approach

 

 

 

--DEC

[PC-Janitor 0]

I'm looking at something similar businesswise.

 

I loved the ease of setting up LTSP, and I have a couple of clients who I have to persuade tommorow to get upgrades of windows, I would much rather offer them Linux.

 

If I were to set them up with LTSP, they could even use an older machine as well, giving them 3 for the price of none.

 

The sticking point of course is Support, I mustn't leave clients dependent on me. I've thought about VNC'ing or X'ing into their boxes for updates, but that's a little risky.

 

I don't think they are particularly willing to learn a new system, although one of them is quite new to computing, their past exposure has been Windows at the office and they've made up their mind that they would like to be able to continue that learning at home - understandable, I wouldn't want to confuse their return to work period with an intro to OpenOffice.org!

 

So in small numbers, the answer seems to be no, but if we can persuade larger companies to upgrade to a LTSP type setup with Crossover office co-installed with OpenOffice.org, the support would be

[*] worth the time spent, because others will be able to re-enforce lessons learned, thus making it cheaper to provide.

and

[*] better accepted due to an overall change of culture within the organisation.[/list:u]

 

I've been playing with the redhat/fedora branch K12LTSP and wonder if it's time for a business branch in the same vein within the same 'fraternity'.

 

Jon

 

P.S. Like your site