I appreciate all the comments in this thread so far. <br><br>Here's a couple of clarifications I would make:<br><br>a) I'm not designing any applications for this research group. Just getting them setup with an environment. This will involve buying some machines and installing applications. But its not something I can spend all, or even a majority of my time over a few months on. Its a side project. I can put in big chunks of time, but I can't sustain that.
<br><br>b) In fact, the head of the lab is in engineering. He has picked Unix over Windows for his environment. But he's not a person who can or should setup and administer it over the long run. This is a very flexible, results oriented environment. We don't have any 'stupid' or 'wheel spinning' administrators. And there's no existing culture to mesh with. Starting from scratch.
<br><br>c) I've run linux (or one of the BSD's) on my personal computer for a long time. But I've never had more than two computers that I had root access on. <br><br>So that said, again, I just like hearing everyone's perspective, and its an interesting discussion.
<br><br><div><span class="gmail_quote">On 4/20/07, <b class="gmail_sendername">Douglas Green</b> <<a href="mailto:diego96@mac.com">diego96@mac.com</a>> wrote:</span><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
I have some experience with this... but I'd say up front that a researcher's needs will depend (to a LARGE extent) on what kind of work he's doing. I've found that the engineering/math culture is MUCH more linux friendly than say... pathology (where I worked). John, if he's still around, would probably agree. That said here's my retrospective:
<br><br>I was at Pitt, and found it very very difficult to avoid the use of microsoft. For the sake of efficiency, I'd recommend making sure that he's using whatever OTHER people are using in his computing environment for end-user use. I can feel the "boooos!" already, but this was my experience. There's nothing worse than sending a .pdf or plain text file to some administrator who can't get their computer out of first gear. Another major hassle was Powerpoint- Pitt would ONLY accept poster submissions in powerpoint format. Lame! Bottom line, be prepared to encounter a very non-flexible enviroment IF that's the culture of your department (as it was in mine).
<br><br>Second- I would have liked to keep EVERYTHING in as simple a format as possible. References, primary data (of course), and even emails should be kept in non-proprietary formats. If I did it again, I'd start with a very simple MySql or something, and build on it as needed so that I could cross reference things. I'd get a network storage device, preferably one that's wireless and that I could easily relocate (when they move your office). Backup backup backup!! My laptop was stolen from my car, and I only lost 1 week of data thanks to backups! This should ideally be done on a daily basis (scheduled task to a different hard-drive).
<br><br>While it's just my opinion, I've found that the records I refer back to are those kept in the most durable formats. I have no idea where my old EndNote references are anymore (they've been lost with revisions to the software), but I've got a huge crossreferenceable database (in plain text format) that I use with BibDesk.
<br><br>Best of luck!<br>-Doug<br><br>On Friday, April 20, 2007, at 01:08PM, "Christopher DeMarco" <<a href="mailto:cmd@alephant.net">cmd@alephant.net</a>> wrote:<br>>On Fri, Apr 20, 2007 at 10:30:42AM -0400, Michael H. Semcheski wrote:
<br>><br>>> I've often thought 'boy, if we were starting from scratch, I would make sure we<br>>> never did this, and ...'<br>><br>>I think it's pretty much a given that whatever plan you start with will
<br>>be amended down the road; whatever technologies/implementations you<br>>start out with will eventually be discovered to suck in some aspect or<br>>another. The only thing that can be said with certainty is that "If I
<br>>were starting from scratch, I'd make sure to think things through and<br>>formalize the plan and provide alternatives for future use." You *will*<br>>be changing things later, build with as much modularity and spare
<br>>capacity as can manage, so that later it won't be as much of a pain to<br>>rip out / migrate away from / scale up. The only constant is change, so<br>>plan for things to change and you'll be alright.
<br>><br>>Agility and thoughtfulness are more important than any specific<br>>technologies or configuratia. IMHO.<br>><br>><br>>--<br>>Christopher DeMarco <<a href="mailto:cmd@alephant.net">cmd@alephant.net
</a>><br>>Alephant Systems (<a href="http://alephant.net">http://alephant.net</a>)<br>>PGP public key at <a href="http://pgp.alephant.net">http://pgp.alephant.net</a><br>>+1-412-708-9660<br>>_______________________________________________
<br>>wplug mailing list<br>><a href="mailto:wplug@wplug.org">wplug@wplug.org</a><br>><a href="http://www.wplug.org/mailman/listinfo/wplug">http://www.wplug.org/mailman/listinfo/wplug</a><br>><br>><br>_______________________________________________
<br>wplug mailing list<br><a href="mailto:wplug@wplug.org">wplug@wplug.org</a><br><a href="http://www.wplug.org/mailman/listinfo/wplug">http://www.wplug.org/mailman/listinfo/wplug</a><br></blockquote></div><br>