1 |
>>> Can you rely on Xorg devs to ensure that they are not going to break your |
2 |
>>> multiseat system in the future? |
3 |
>> |
4 |
>> Maybe I'm wrong, but I don't know why there would be (much) more |
5 |
>> likelihood of regression with Xorg multiseat than with anything else, |
6 |
>> including LTSP and all of its dependencies. |
7 |
> |
8 |
> Because fewer people are testing it. |
9 |
|
10 |
That's fair but Gentoo makes it easy to roll back if necessary. |
11 |
|
12 |
> You can get low-powered Linux systems for $100 or $150 - either a little MIPs ShivaPlug or (I guess) a secondhand atom nettop (Acer Revo). |
13 |
> |
14 |
> If you save 2 hours per machine by using a standard and common thinclient configuration, then the hardware has paid for itself. |
15 |
|
16 |
You're saying use built-in thin-client firmware (on the SheevaPlug for |
17 |
example) along with something like VNC or NX on the server to save |
18 |
time over an LTSP setup? That would mean giving up some software |
19 |
control. Assuming multiseat works, is there an advantage to this over |
20 |
multiseat? |
21 |
|
22 |
> If you have to employ a Linux sys admin to help you fix a complicated problem with Xorg multiseat, then it will run you at least $100 or $150 for those 2 hours. That's how you should be valuing your own time, too. |
23 |
|
24 |
LTSP and its host of dependencies seem much more complicated to me |
25 |
than multiseat. |
26 |
|
27 |
- Grant |