jhnc: (Default)
This Is My Life, Rated
Life: 8.1
Mind: 7.9
Body: 8.2
Spirit: 8.5
Friends/Family: 4.1
Love: 6.9
Finance: 8.5
Take the Rate My Life Quiz
jhnc: (Default)
If/When I get back to posting anything, it'll just be on dreamwidth. No more crossposts to LJ.
jhnc: (Default)
I was considering purchase of vbulletin but it appear impossible to sensibly keep to the terms of the license:

https://www.vbulletin.com/order/license_agreement.php
 
Section 2 states:

You may not give copies to another person, or duplicate the Software by any other means, including electronic transmission. You may make one copy of the Software in machine-readable form for backup purposes only

This appears to mean:
  • I would not be able to install it on a webserver after I have downloaded it, as this is duplicating but not for backup purposes.
  • In fact, I would never be able to transfer the software to any other storage media, even on the same machine, as it is hard to see how to avoid two copies existing during the "move" operation.
  • I could not install it on any server where I actually care about not losing my information. Who in their right mind only keeps one backup of their operational data?!? Even ignoring the fact that for disaster recovery I have three levels of backup (host, onsite, offsite), I would still have to disable both RAIDZ2 and filesystem snapshotting on my servers.
They should consider redrafting the license to be less ludicrously onerous.

The Borland "like a book" No-Nonsense License from the '80s was a model of clarity and good sense.

you must treat this software just like a book, except that you may copy it onto a computer to be used and you may make archival copies of the software for the sole purpose of backing-up our software and protecting your investment from loss.


jhnc: (Default)

I want to calibrate my LCD monitor but own neither spectrometer nor colorimeter. I know very little about the theory or terminology.

Presumably one of the purposes of calibrating is to make the monitor display colours realistically. In other words, if I took a photo of an object in the real world and displayed it on my monitor while holding the object beside it, I assume the ideal would be to have the colours on the monitor resemble the colours of the real-world object.

If they match, I'm set. If not?

I have a webcam (Microsoft Lifecam Cinema - semi-HD, glass lens, manual focus, great for barcode scanning). I could set it up so that it can see both the original object and the monitor display. I could then map the colour differences between the real object and the onscreen representation. Do this for several different colours and perhaps a simple best-fit linear transform could be calculated. (Or a lookup table.)

Something like:

  1. Obtain test target (sheet with several solid blocks of different colours?)
  2. Point webcam at target, grab a still, display onscreen
  3. Point webcam at target+screen, grab another still
  4. Match up corresponding points of target and screen
  5. Calculate transform from screen to target
  6. Stop if no adjustment required
  7. Apply transform to monitor setting
  8. Repeat from 2

Might this work? What obvious flaws have I missed?

jhnc: (Default)
This Is My Life, Rated
Life:
7.9
Mind:
8.2
Body:
9.2
Spirit:
8
Friends/Family:
3.6
Love:
4.6
Finance:
8.4
Take the Rate My Life Quiz
jhnc: (Default)
Testing, testing, one, two, three, four.

Meme

Dec. 26th, 2008 04:05 pm
jhnc: (Default)
List the towns or cities where you spent at least a night away from home during 2008.
Mark with a star if you had multiple non-consecutive stays.

Birmingham
Broadmayne
Bangkok *
Beijing
Chiang Mai

Apparently I like places that start with "B".

jhnc: (Default)
I'm finally getting around to switching from an out-of-date version of galeon to firefox. Now that I've managed to get vertical tabs working and tweaked a whole bunch of settings, it seems mostly okay. The bookmark organiser/sidebar system is completely alien though, and with forty or fifty tabs open in a window it's a noticably slower at switching than galeon with the same number.

Am also experimenting with the RSS capability and am curious to know why all the lj rss feeds I've added contain the full content, except for autopope's, whose feed rather unhelpfully only contains titles.
jhnc: (Default)
What would be a good choice for a hostname for a videoconferencing system?

Clearly, "videoconferencing" would be the most obvious and precise but it is rather too long.

Things like "talk", "chat", "confer" don't suggest any visual aspect to the communication. On the other hand "video" alone is too general - we already make "vodcasts" (or whatever the cool name is these days) available from an entirely separate location. Contractions like "vidcon", "vconf", and the like look rather ugly and aren't really very obvious.

"MCU" might be appropriate for the multipoint control unit (cf. smtp, mx, pop, ftp, etc), but is completely non-intuitive unless someone is already in the know about terminology and another name is still needed for the endpoints. "VTC" (videoteleconferencing) doesn't seem to be an especially widely used abbreviation.

I wonder what do other places use? The only example I've found in a brief search is "video".
jhnc: (Default)
A poll of five people P1..P5 is undertaken.
They must rank seven choices A..G.

The answers are:
 ABCDEFG
P14137526
P21324657
P37125436
P47125436
P51325647

The mean of the ranks can be calculated and a global ranking assigned.
Should any significance be given to it?

In particular it appears that B and C are ranked highly and G very low.
Can it be assumed with any confidence that this represents any kind of consensus and that the result is not in fact simply random?

I did some googling and the Friedman test appears to be appropriate and perhaps one could post-process with a Bonferroni-Dunn post-hoc test? However all this is way beyond the statistics I did two decades ago and have long since forgotten.

Help?

jhnc: (Default)

The answer is not "Bangkok" apparently.

While ascending to the highest point of Thailand a few weeks ago, I was informed that only foreigners use "Bangkok", and that to any Thai the correct name is "Krungthepmahanakhon Amornrattanakosin Mahinthrayutthaya Mahadilokphop Noppharat Ratchathani Burirom Udomratchaniwet Mahasathan Amornpiman-Awa-Tansathit Sakkathatiya Witsanukamprasit".

Coincidentally, this question was asked on QI last night, but sadly Stephen Fry didn't attempt to pronounce it, merely giving the translation of the first phrase ("city of angels") and pointing out that the Guinness book of Records lists it as the longest place name in the world.
jhnc: (Default)
I noticed this sign in the Beijing International Conference Center, on the floor where the Go part of the 1st World Mind Sports Games took place:

jhnc: (Default)
Statistically, the players ares grouped into four band strengths: professional, high dan, low dan, kyu. I have some hope against the latter two, little against the second, and none against the first. The ratios are about 1:2:2:1.

My first round opponent appears to be Chang Li-Yu, a 7 dan professional from Chinese Taipei.

I shall enjoy my breakfast and attempt not to play *too* embarrassingly.
jhnc: (Default)
Free wireless! But tunnelling X over SSH is painfully slow. Shall have to set up VNC. It has taken five minutes for the screen to update and typing blind is tedious.
jhnc: (Default)
I'm taking a break from packing.

At 5am, I head off to Heathrow to fly out to Beijing as part of the twenty-two member Go team representing Britain in the 1st World Mind Sports Games. I don't think there has been much mention in the UK press, so if anyone sees anything, please comment here.

I believe I can get net access from my hotel so, unless the charges are extortionate, I may "blog" while I'm out there. As I have 32GB of SD camera storage, there may even be pictures eventually. My fingerprinty new Eeepc 901 will get its first proper trial (network access terminal and Go database).

Apparently I was wrong when I guestimated that Chiang Mai would be low/mid 20s. It seems to be up in the 30s at the moment. Beijing is, however, in the low 20s, so I should survive the ordeal of wearing a jacket for the opening ceremony in the "Beijing Olympic Sports Center Indoor Stadium" (whichever one that is). So, the raincoat will stay at home.

Back to packing.

cords

Sep. 23rd, 2008 10:54 pm
jhnc: (Default)
Getting on for ten years ago, I bought a pair of cords and hung them over the back of a chair unworn. More unused clothing gradually accreted. Several years later, having completely forgotten about them, I discovered the trousers still there, and discovered that they were now too small. Oops.

Looking for a suit bag just now, I investigated a storage box and discovered the cords (as well as a few unused adminspotting t-shirts (which coincidentally reminded me that someone emailed a few days ago to ask if I knew where to obtain any. I guess I do...)). Due to my accidental crash weight loss in January, the trousers now fit. Yay!
jhnc: (Default)
Observation: Lynx "snake peel" shower gel closely resembles Tufanega.
jhnc: (Default)
Despite only being ranked 83rd in the UK, I have been invited to join the British Go team at the First World Mind Sports Games in Beijing in October, to take part in the Men's Individual event.

Professionals may enter, so I have precisely no chance of winning and, given my results in the British championship over the last few years, I'd be surprised not to come approximately last (but I suppose someone has to...).

I am in two minds whether to accept: Ego boost of having taken part and ability to put "represented UK in ..." on my CV versus humiliation of losing all my games and the knowledge that much stronger British players didn't take part (why not?).