Forum OpenACS Q&A: Somewhat OT: Bootcamp revisited

Collapse
Posted by Walter McGinnis on
When Bootcamp was started reading Edward Tufte was "required" (see bottom of this page: http://web.archive.org/web/19991122033148/photo.net/teaching/boot-camp.html), but IIRC a lot of people skipped it.

Here's an article in NY Times about some of his recent work that takes MS Powerpoint to task: http://www.nytimes.com/2003/12/14/magazine/14POWER.html?ex=1071982800&en=799ad449b398c2d7&ei=5062&partner=GOOGLE.

It might be interesting to look at the mentioned 28 page pamphlet if we every redesign or extend wimpy point.  It actually might be worth reading for general UI improvements to the site.

Collapse
Posted by Walter McGinnis on
And here's the link to the Tufte's essay: http://www.edwardtufte.com/tufte/powerpoint.

The link is holding up pretty well, considering its now mentioned on ./, of course that may have more to do with the big news of the day.

Collapse
Posted by Carl Robert Blesius on
http://uptime.netcraft.com/up/graph?site=http%3A%2F%2Fwww.edwardtufte.com%2F

Which version of OpenACS is he running?

It is still on the front page of Slashdot and breathing (even though he has a lot of "visual information" on his site).

Is there a way to check?

Collapse
Posted by Walter McGinnis on
IIRC, Jin S. Choi built his site on top of ACS 3x, but maybe it was earlier than that.

I emptied my cache, and just tried that link from the /. front page, it came up in about a second.  The front page of the site is fast as well.

Your best on hits info is to inquire directly.  They do mention Jin and ACS with a link to openacs.org.

Collapse
Posted by Patrick Giagnocavo on
I seem to recall being told by someone that it was 3.x with Oracle.
Collapse
Posted by Janine Ohmer on
Yes, Tufte's site is based on ACS 3.4.9, according to the version history file.  It uses Oracle, and is running on an average Redhat 7.3 system.

I don't know the early history of the site, but DVR (David Rodriguez) is the primary maintainer now.  furfly just hosts the site.

Collapse
Posted by Walter McGinnis on
Any word on hits and average response time while /. featured the site?
Collapse
Posted by Mike Sisk on
Ok, I got permission to share the traffic and hit counts from Edward Tufte's site during the Slashdotting. The 13th is the day before the site was mentioned on Slashdot is represents their average traffic.

13/Dec/2003 Hits:  68.7k, Traffic:  1.10 Gigs
14/Dec/2003 Hits: 457.8k, Traffic:  7.81 Gigs
15/Dec/2003 Hits: 436.8k, Traffic:  7.41 Gigs
16/Dec/2003 Hits: 228.4k, Traffic:  3.92 Gigs
17/Dec/2003 Hits: 209.2k, Traffic:  3.60 Gigs
18/Dec/2003 Hits: 155.7k, Traffic:  2.49 Gigs
19/Dec/2003 Hits: 135.5k, Traffic:  2.08 Gigs
20/Dec/2003 Hits:  52.2k, Traffic:  0.82 Gigs

This isn't an especially large event, but it's interesting in that the event lasted much longer than a typically Slashdot event. On other sites we host that have been slashdotted, the traffic has been a bell-curve shape with a single peak.

Here, the peak traffic on the 14th and 15th was at around 2 PM EST and  had a substained data transfer of nearly 4 Mbit/sec.

We don't monitor response times, but the site never went down and the load on the server never exceeded 4. As luck would have it, this is our least powerful server -- it's a Dell 2450 with a single 733-MHz PIII CPU and only 1-GB of RAM. It has 4 18-GB SCSI drives on a hardware RAID 5 controller. It's a shared server and hosts a few other sites in addition to Tufte's.

Collapse
Posted by Jon Griffin on
Mike,
Would it be possible to share this on the OACS website under why AOLserver? Or something to that effect?
It would be a nice thing to point to when people ask why they should use AOL.
Collapse
Posted by Mike Sisk on
<blockquote><i>Would it be possible to share this on the OACS website under why AOLserver?</i></blockquote>

Certainly -- not a problem. I got permission to share this data publically.

Collapse
Posted by Andrew Piskorski on
Heh, that makes for the start of a nice little case study, Mike!