Je hebt het gevoel in een reclamespot terecht te zijn gekomen, en de lokale overheid laat geen moment onbenut om je ervan te overtuigen dat je hier moet zijn als je in China wilt komen investeren. De bijbehorende glossy folder vermeld: electriciteit: 8 eurocent per kwh, drinkwater 31 eurocent per kuub, overal glasvezel aanwezig voor supersnel internet, labor cost…. Labor cost? De folder meldt dat een “gemiddelde” programmeur hier 2300 euro per jaar kost, en voor een software engineer met master’s degree moet je gemiddeld aan 6600 euro per jaar denken. Je moet hier dan nog wel 31% sociale premies bij optellen als je de werkgeverslasten wilt berekenen. Ook wordt terloops nog even vermeld dat zwaar chemisch afval of radioaktief afval bij de overheid kan worden ingeleverd indien je daar gebruik van wenst te maken. Verder een stapel folders van kale bouwkavels (“removal and demolition are under way” – in China schildert de overheid een teken op je muur en dan heb je 14 dagen voordat je woning tegen de vlakte gaat. Voordeel is wel dat je altijd andere woonruimte krijgt), panden in aanbouw of compleet gerealiseerde kantoorunits in elke gewenste grootte die je maar wilt. zie cdht.gov.cn
Daarnaast krijg je de beschikking over iemand van de overheid die je full time komt helpen bij het verkrijgen van alle benodigde vergunningen, als je je daar al zorgen over maakt want in de folder staan alle stappen die voor de vergunning nodig zijn in een overzichtelijk schema vermeld. Kortom: te mooi om waar te zijn. Denkend aan ons kikkerlandje waar de overheid de ondernemer op elk mogelijke manier het leven zuur probeert te maken is het toch wel een verademing, zo’n dictatuur. Als proef op de som zijn we dezelfde dag ook nog op bezoek geweest bij Plenware, een fins detacheringsburo werkzaam in de ICT sector. Zij hebben net een Chengdu office geopend, en zijn op het moment dat we binnenkomen net klaar met inrichten. Op de vraag of alles inderdaad zo gemakkelijk gaat kwam het antwoord dat we al min of meer hadden verwacht: sommige dingen gaan makkelijk, andere moeilijk. De ervaring was dat vaak de simpele dingen lang duren, bv het regelen van een mobieltje. Dat komt ook overeen met onze ervaring toen we een paar batterijen wilde kopen en een half uur aan het onderhandelen waren met de beste man omdat hij vond dat we teveel betaalden. In China gaan makkelijke dingen moeilijker, moeilijke dingen gaan gemakkelijker.
Al met al ben ik zeer enthousiast geworden om hier iets te gaan starten. In Nederland wordt het steeds moeilijker om software te ontwikkelen, goede mensen zijn niet te krijgen, en hier in China lijkt de kans op succes veel groter. Na eerdere mislukte pogingen om software ontwikkeling in India en in China door een outsourcing bedrijf te laten uitvoeren denk ik er nu over om hier een kantoorunit te huren en een development club te starten. Ik ben om.
Ronald van Woensel - Chinareis2008.nl
[ add comment ] | permalink | related link |
( 2.9 / 329 )During my trip to China a few weeks ago I met Ernst Fuld again. Ernst is CEO of the NGN, the dutch IT-professional platform. He told me that he has tried to test some usb-sticks and see if they could be written and erased so many times that they break. As you probably know, flash memory can only be written a limited number of times. Each time you erase (rewrite) the memory cell, there is a slight chance that it breaks. Normally, the memory is guaranteed to last for 100000 rewrites, but has anybody tested this on a regular usb stick?
There is a problem however. The sticks are known to use a technique called "wear levelling". The data you write to the stick is remapped to memory area's where the number of rewrites is the lowest. This means that you cannot test 1 cell by re-writing a small file over and over. You have to fill the WHOLE stick in order to rewrite all cells. And even if you manage to break a cell, there is also error-correction logic and bad-block remapping that will correct the error without notice. So the flash memories seem to be unbreakable.. or not!?
That's why I wrote this simple app: Flashkiller. It's a free download so you can test the reliability of your memory stick yourself. If you broke a usb-stick or memory card with it, please respond to this post!
May 2 update: the kingston 512mb is now @19000 rewrites and still without an error..
june 13 update: The kingston is DEAD. After 65000 writes it becomes a read-only stick. Nothing can be written any more, and when i try to write it gives an error!!!!
[ add comment ] | permalink | related link |
( 3 / 364 )I'm using firefox as my default browser, but for testing and development i need explorer every once in a while. A couple of weeks ago, my microsoft update asked my to upgrate to IE7. So why not? well, always reluctant to install microsoft stuff, but if everybody uses IE7, i should at least test websites for this new browser.
So i installed IE7. What a bummer! I got an icon, but that was about it. Clicking the IE7 icon shows the hourglass for a fraction of a second, and then it exits immediately. So i uninstalled it and installed it again. Nope.. Then I reinstalled IE6. Works fine!
Next step: google the answer. Googling for "IE7 wont start" gives you this link:
http://support.microsoft.com/default.as ... ;spid=8722
Wow! microsoft already solved my problem!..but n o.. this wont fix it.
of course, further searching the microsoft site using their own search function never gives a valid clue.
so.. now what? Runing IE7 without plugins also doesn't help. It just wont start.
uninstalling .net 1.0, .net2.0 visual studio and a lot of other apps? no luck.
So after a while I stumbled upon SFC. This utility is able to check *ALL* critical system files and reinstalls them if neccesary,
Running SFC /SCANNOW went well. I took about half an hour, and you only see a progressbar, no information, no logging and at the end, the program just exists without any report.. BUT it did the job: IE7 now starts. It took me about 6 hours to get IE7 working, but in the end i finally can test this great piece of work from microsoft.. Why can they *never* get their software right?
[ add comment ] | permalink | related link |
( 3 / 470 )How often does this go wrong: multiple dhcp servers on your subnet? Because routers, accesspoints, modems boxes ALL have dhcp servers nowadays, a double or triple dhcp server in your network is not uncommon. But tools to identify which dhcp servers are available are not easy to find.
To identify that something might be wrong, you can use ipconfig on a windows cmd box:
C:\Documents and Settings\rvw.KOEKJE>ipconfig /all
Windows IP-configuratie
...
Ethernet-adapter LAN-verbinding:
...
Standaardgateway. . . . . . . . . : 10.237.147.2
DHCP-server . . . . . . . . . . . : 10.237.147.50
DNS-servers . . . . . . . . . . . : 213.51.129.37
213.51.144.37
etc
This shows you your CURRENT dhcp server, but if there are multiple dhcp servers running by accident, you will get an IP address from a RANDOM server!
Same with NETSH. The command NETSH DIAG SHOW DHCP only shows your CURRENT dhcp server and does not do a scan.
Even utilities that claim to do this, only show one or only the current dhcp. DHCPexplorer from nsasoft is such a fake application.
Fortunately there is a tool that comes standard with windows, called DHCPLOC. If it is not installed, you can find it on your windows CD under the \support\tools directory OR get it from
here
Give your current IP address as an argument to dhcploc, and press "d" (discover). Press d a few times to get a good view.
C:\WINDOWS>dhcploc 10.237.147.101
15:28:30 OFFER (IP)10.237.147.199 (S)10.237.147.50 ***Form
15:28:32 OFFER (IP)10.237.147.170 (S)10.237.147.1 ***
15:28:32 OFFER (IP)10.237.147.199 (S)10.237.147.50 ***
15:28:34 OFFER (IP)10.237.147.199 (S)10.237.147.50 ***
15:28:35 OFFER (IP)10.237.147.170 (S)10.237.147.1 ***
15:28:35 OFFER (IP)10.237.147.199 (S)10.237.147.50 ***
15:28:36 OFFER (IP)10.237.147.170 (S)10.237.147.1 ***
15:28:37 OFFER (IP)10.237.147.199 (S)10.237.147.50 ***
There! there is a rogue dhcp server running at IP 10.237.147.1 !!
So you know.
(the rogue dhcp server turned out to be a netopia dsl router, which only changes dhcp server settings after a REBOOT! Shame!)
Edit:
Another nice tool is DHCPFIND.EXE, from roadkil. Its free and works, and is a GUI tool, not a commandline tool. Only problem is that it does not show the dhcp server IP, so dhcploc is still the best utility for the job. If you want to try dhcpfind, look here
[ add comment ] | permalink | related link |
( 2.9 / 956 )Already for a few years, i'm running MSERV, a great opensource mp3 jukebox which is completely client-server and has a good rating- and genre system which allows it to play random music that corresponds to the current mood, current lister etc.
You can rate numbers as superb, good, normal, bad, awful etc. and it takes this into account to play numbers more or less often, depending on your rating. Everybody can rate, so this is a good jukebox to use with a group of people.
There are a few clients available, and I've taken MservClient as a base to further develop the windows client. MservClient is written by Marco Schulze and Marc Klinger, and can be found here. Because I've changed the whole concept of MservClient, I adopted a new name: Mserv4win. Also, there is not much left from the original code of mservclient. Mserv4win takes the look and feel from winamp and puts it on top of mserv.
One of the things missing in Mserv is a good "cover picture" feature, so I've also added an "x cover" command to get a cover pic from the server. The client still talks to the original server that does not support this command, but with this feature, the experience is a lot better.
Here's a screenshot:
;)
Download Mserv4win: MServ4win.zip
Download source (delphi7): sources.rar
If you want to patch your mserv for "cover" support, use this to patch mserv's cmd.c and recompile: cmd.patch
or replace cmd.c with this version: cmd.c
make sure you put cover .jpg files in the same directory as the album mp3's. Also, make sure the jpg is not too big (max 50k or so)
[ add comment ] ( 75 views ) | permalink | related link |
( 2.9 / 1291 )Back Next
Calendar
