It’s a MiniITX case with a DC-DC powersupply and room for a slim DVD and a PCI-card. It’s also got room for a nice and quiet 3.5″ HDD (Say, a Samsung Spinpoint P120).

However there is no active cooling provided, which could prove a problem for the HDD.

The problems occur both towards the linux-server running samba 3.0.x and towards the stationary windows box. For the client it’s registered as either a “Delayed write error” or as “Disk Full” or whatever it comes up with when it fails towards the samba server if it’s running with “oplocks = no” which I turned on because it reported “request_oplock_break” in the log.

Using wired interface I can do stable transfers without any issues, I currently get 8MB/s towards the linux box with it’s 100Mbit network card, which is just about what that interface can take.

I’ve looked at the network traffic using Ethereal and it displays TCP DUP ACK, TCP Retransmission and once now and then even a TCP Window Update. The problems during transfers seem to be caused by TCP DUP ACK and TCP Retransmission.

I’ve tried turning off all the bells and whistles on the Router but it’s still running unstable. Yesterday I opened up a support case towards D-Link which resulted in them telling me to DOWNgrade(?) the Firmware from version 1.6W which it was delivered with from the factory to firmware version 1.6 which is the latest version offered by D-Link Sweden. I tried that earlier this evening with little success. The up(down)grade of the firmware went smoothly and everything came up as it should. But sadly enough I was able to reproduce the problems again…

I think I’ll try replacing the network cables just to be on the safe side… apparently there is something called CAT6 cables around now, might aswell upgrade to that.

I personally haven’t used Bryce in ages, but I look forward to testing it again. Especially to put it up to a comparison with Vue 5 which I recieved a license for when I bought Lightwave earlier this year.

The new one is made by Hirsch and part of a series they call Professional. The reason for the delivery time is that it had to be specially ordered since the Oris Artelier uses the non-standard width of 21 mm.

I choose to get it with a clasp locking mechanism, which I’m very happy with. It’s much smoother to take it on and off now and I think it will also make the wristband last much longer than the old one.
The Oris Artelier Big Date is a classic swiss-made mechanical automatic with a twist, it features a double digit date and a non-centric seconds-hand in a very nice 30 jewels movement based on ETA 2892A2.

The truth of it is that they really did use them but the brand later (almost) died out to resurface in 1998 when a small Milanese publisher brought it back. Today it’s a very widespread product known for it’s quality and used by many authors etc.

Made with bound cardboard covers including an index ribbon aswell as a elastic band to hold it closed. The pages are acid-free and archival quality, they’re thin but hold up well when used with a fountain pen. I’ve seen some feathering issues with my Swisher ink. But nothing troublesome. I will try with Pelikan or Lamy ink next and see if they work better.
Moleskine have a range of books in several sizes with both plain, ruled and squared pages. There is also special models aimed at artists who use watercolors etc.
Even though the moleskine books are nice, the ultimate notebooks must be the ones offered by Renaissance Art. Hand made leatherbound books made the same way that they were made several hundred years ago.

There is also a possible combination option, Renaissance Art also make wrapping covers for the moleskine notebooks. Prices range from $20 to $39. Personally I’ll stick with plain Moleskine for a while, however a book cover from Renaissance is very tempting (I like to be careful with my books, even the paperback ones).

It’s designed to simulate the feeling of walking barefoot on soft natural materials, such as sand or grass.

Todays society has too many unnatural and hard materials that our skeleton and joints does not appreciate.

I tried out their sandals called Kisumu with the intention of possibly using them at work. Today I use sport sandals (currently Nike) since they are comfortable and airy, however my current pair are getting fairly worn out by now.

The MBT sandals were nice and soft giving a nice bouncy feeling when walking. Due to their rounded soles they give a forward rolling motion much similar to what some running shoes provide (e.g. the Reebok Premier Road Lite series). Although they don’t support the foot like normal shoes it still didn’t feel strange. But I noticed that some of the others who tried hem had some difficulties, likely because they have grown accustomed to the artifical support from other shoes.

Before closing I just have to show these two pictures visualising the difference between normal shoes and the MBT-series.

The only downside I’ve found is that the price of these anatomical and medically approved etc. shoes is (like all other specialised products) higher than normal. I’ve seen prices around 180 - 210 euro.

I’m planning on getting a pair of black Kisumu for work (see the picture above). I think they will help my legs recover from the daily commuting and also strengthen and improve my posture.

I haven’t been able to find any reliable reviews of this unit, but on paper it looks very interesting for everyone who has heaps of mounted slides to scan (doesn’t almost everyone?).

I’m sorry to say that I will have to redo almost all of the layout for this site. I made the mistake of using an existing layout as the base when redoing it. And the clutz that set it up has made a total mess of both the php and the CSS. Especially the CSS is troublesome since the same topics appear in multiple places and entried override eachother.

The uploaded file could not be moved to .

The message above is displayed if there is a problem uploading or using the uploaded file, this is normally associated with security issues, such as the user that the webserver is running under doesn’t have write permission in the folder.

There is a thread on the wordpress site that discuss this issue. The most common cause for this problem is invalid directory permissions.

My problem however was not solved by changing the permissions of the folders. No matter what I tried it refused to work.

After turning to google I found out that this type of problems may also be caused by PHP being configured to run in SAFE MODE safe_mod=On in php.ini) which has the same effect as incorrect permissions. Php will refuse to read files that are not owned by the same user as the script being executed.

So by turning off safe mode (safe_mode=Off) the file uploads worked without problems… but personally I prefer running with safe mode enabled. So I needed to find another solution to allow file uploads even though safe mod is enabled..

It so happens that there is a related configuration option in php.ini. By setting safe_mode_gid=On safe mode checks the files GID (group) instead of the UID (user). When combining this with a sticky flag on the uploads folder (and it’s underlying year-folders if you have that option is enabled in wordpress) we can get file uploads to work properly even though safe mode is still enabled. Here is a list of what to do:

• Enable safe_mode_gid=On in php.ini
• chmod 7777 wp-content/uploads
• If the organisation option is enabled in wordpress, create one or more year folders and set chmod 7777 on them aswell.
• Restart the webserver so that the php.ini changes take effect.
• Try uploading a file.

Hope this helps. Good luck!

For those of you who like regular expressions (regexes for short) here is my take on a regex to parse a ISO-date into it’s components. It supports date-only, date & time, date & time +timezone and both short and long formats may be used (it’s possible to exclude the separators.

The expression below should be used as a single row. The first row matches the date-part, the second and third which are optional, maches the time and then there is the also optional, timezone pattern.

^(([0-9]{4})-?([0-1][0-9])-?([0-3][0-9]))
[ T]?(([0-2][0-9]):?([0-5][0-9])
(:?([0-5][0-9])([,.][0-9]+)?)?)?
((Z)|(([+-][0-2][0-9]):?([0-9]{2})?))?

Please not that this should not be used for ISO-8601 validation since it doesn’t support the week or day of year notations. It’s also not strict regarding “valid” months and days of month. It’s possible to enter month 18 for example. I choose to leave that out since I only needed to parse expressions and didn’t want the regex to be longer than necessary.