OpenOffice.org 2.2 released

Posted Mar 30, 2007 6:44 UTC (Fri) by drag (subscriber, #31333) [Link]

It's just about the same thing as going (on a much reduced scale for MS Office to OO.org conversion) to some Windows-only programming shop, ripping out all their Windows servers and Visual basic and replacing it all with Eclipse or Vim or whatever you prefer and walking away.

That's absolutely batshit insane and it would never work.
But I don't feel (and I doubt you feel like this either) that a Linux-based workstation is a much inferior solution for programming/development platform then a Windows desktop.

Generally how this works is that if your going to introduce new applications or desktops into a workplace that you have at least one guy present that knows enough about it that they can perform what duties that application needs to be used for.

Then that one person sets up a training date, maybe a evening or two, were you take the more tech-savy people from a few different departments are trained in how to use it.

Then you send them back and put those people in charge of answering questions on how to deal with that. And if any more advanced questions pop up then you (or whoever) can figure out how to deal with them.

And, seriously, how many folks in your place need help with VB scripts in OO.org?

Also you buy documentation, books, whatever for the application and make sure that people know were to find them.

I've worked in a community college were I had to help out with tech support for a few months. I'd have to wander around and help students (and teachers) out. We'd had to deal with migrating from OS 9 to OS X for a lot of people. Had to deal with going from quark express to indesign. Going from Mac-based desktops for 2d design to show people some about Windows so they can use the 3d workstations. STuff like that.

The biggest problems we faced on a regular basis was that the Windows file servers performed so poorly that the IT people had to keep changing stuff around in a effort to improve performance. And that would cause changes in how the people using the Macs found their network shares and such.

It's quite a bit different in acorporate environment, but I expect the fundamentals are the same.

OO.org tries to keep very closely with MS Office so it's not _that_ different, but I know it's differnet enough to cause some little problems.

So in summary.
1. Get _somebody_ familar enough with the application so that they know how to use it well enough to get the job done.
2. Have a training date were you have select more tech-ish savy people go over the differences and such.
3. Send them back to their workmates so that they have somebody that can intercept easy questions.
4. Do printouts and provide books and other documentation (internet links, for example) so people are able to find out stuff for themselves.

I figure that for a small and medium sized business that isn't terribly well integrated into Windows structure (as in they aren't depending on AD + Office + Exchange + etc etc for business workflow) to migrate successfully to OO.org.

It'd take somebody's weekend, and then a evening's worth of overtime for a half a dozen people for the training session.

Now I bet you can get buy with much less, but this approach I feel will provide a much higher likelihood of success.

Posted Mar 30, 2007 1:31 UTC (Fri) by dskoll (subscriber, #1630) [Link]

Printers and scanners have never been a problem. We have three printers (an HP LaserJet, an HP all-in-one fax/scanner/printer and a Samsung printer) and they all just worked out of the box.

Naturally, it's easier to start running Linux than to switch from Windows --- it's called Vendor Lock-In. We switched from a proprietary PBX to Asterisk only after our irritation with the proprietary PBX drove us to the breaking point. My guess is that Windows lock-in is far more powerful than PBX lock-in, so there's much more difficulty in changing.

But that's OK. I see our use of Linux as a competitive advantage, and if our competitors are stuck running Windows... :-)

I hear your complaints about the "little" problems that add up and are easier to solve. The best way around those is to avoid the problems in the first place -- make sure you buy Linux-compatible hardware and choose Linux-friendly vendors. And the "cheap" solution of switching to Windows will have many other costs in the long run.

Posted Mar 30, 2007 1:43 UTC (Fri) by beoba (guest, #16942) [Link]

I've found that, consider that I've never had to manually install drivers for my printer (just select it off a list), I'd argue that things are easier in Linux, I just click a couple relevant links and select my printer from a pulldown list and it works. However, I had to know the above address before I could do that..

There will always be specialized tools for which there is no Linux equivalent. Are those something that we should concentrate on, or should we instead aim for the common things that are useful to the non-specialists? If we get those mainstream areas down pat, I don't see why the specialized tools wouldn't end up migrating of their own accord.

Posted Mar 30, 2007 8:33 UTC (Fri) by nix (subscriber, #2304) [Link]

Even the stuff that's reasonably compatible is problematic sometimes. e.g. I just got a Samsung ML-2250. I picked it *specifically* because it used PCL so would work under Linux, but it seems that it uses PCL6 or something that GS doesn't fully support yet (you can't get full 1200dpi resolution but are stuck at 600dpi). Samsung actually ship CUPS filters that allege to support this, but they're enormous binary-only lumps that are *excellent* at segfaulting. (pxlmono works fine, but you're restricted to 600dpi, again...)

And this is a printer with relatively *good* Linux support.

(I wonder if there are docs for the 1200dpi stuff so I can add support to GS: I haven't spent the time but that's my own damn fault.)