Bad sectors on my laptop hard disk. Luckily, IBM will spring for a new one… I think the battery has developed a “memory” as well, but I’ll play with that more before I complain too loudly…
2.5.59 problem for the day: USB storage doesn’t work for my DSC P3 any more.
Went to 2.5.59 on my laptop. Now I just need to get it to work… First step, get all the modules I need for day to day life compiled and installed. I started with the USB stuff, and now I need to worry about the wireless networking.
The USB hub seems to crash when I remove devices:
[c1520240] link (01520032) element (01521030) 0: [c1521030] link (00000001) e0 SPD IOC Stalled CRC/Timeo Length=7ff MaxLen=0 DT1 EndPt=1 Dev=2, PID=69(IN) (buf=12262ca4) [c1520270] link (01520032) element (01521030) 0: [c1521030] link (00000001) e0 SPD IOC Stalled CRC/Timeo Length=7ff MaxLen=0 DT0 EndPt=1 Dev=2, PID=69(IN) (buf=12262ca4)
Over and over again…
Well, here’s a summary of my weekend. I spent Saturday afternoon hosing down my house and watch flames about 3 – 4 times the height of the trees approaching the edge of the suburb. We didn’t end up having anything in the street burn, but we were listed as a suburb under threat, and we did have burning embers falling out of the sky at one point.
Catherine and Andrew evacuated because of asthma, and me not wanting a pregnant woman and a 11 month old in danger.
I guess we were lucky, because the suburb is new enough to have very few large trees — so gutters didn’t have leaf litter in them and stuff.
My pictures are by no means the most amazing, but these show what an average Canberra person saw on the weekend. Note that I didn’t always have time to take photos — we spent a lot of time with a hose.
It is still raining burnt leaves and at the worst bits of branch and stuff were falling on us. These must have been lifted into the air by convection currents, and must have travelled some distance.
LZW decompression pain related to the STRING+CHAR+STRING+CHAR+STRING special case… This doesn’t seem to be working for me. The hardest part with this is that I don’t know exactly what the decompressed stream I am playing with looks like, so I am not sure what is missing. The next logical step is for me to implement a short LZW compressor (which creates PDF compatible streams), and then try to decompress that and see if the results are right…
This is of course dangerous, because compression is the bit covered by the patents (although those patents don’t exist in my country).
Started thinking about my UDP shell script article. I think I am going to have to write the two hundred word intro to TCP as a first step…
The CVS version of libmplot now supports a quick and dirty raster scaling… Looking at LZW compression because my PDF code needs it. It would appear that the Patent expires mid this year, and no Australian patent was every issued.
Like all good inventions, it’s obvious once you understand it.
Based on http://dogma.net/markn/articles/lzw/lzw.htm, the following string
Where I have used the ’s to indicate code boundaries. The problem from an implementation point of view is that each code entry might flow across a byte boundary. In fact, in PDF 1.3, the codes can be 9 to 12 bits long, depending on the size of the codebook.