Testing ads for malware presence is not as simple as testing an executable for a virus because in general case ad is a combination of Flash, JavaScript and HTML documents hosted on 3rd party servers which may change content of these documents at any time. This means the testing machinery have to do repeat tests all the time but given that there are many creatives in ad network there is a limitation on how quickly you can do repeat tests. If malware author is smart and for example implements a strategy where malware is active only on each 100th impression it may take quite long time to detect the problematic ad.

Doesn't SIM card lock after 3 tries with a wrong PIN code? How do you brute force this?

It is not completely true. If you take picture of me the copyright might be yours but you are often limited in what you can do with the picture. Unless I sign a model release form. As I understand according to USA law there are some cases when non-private use is allowed without the signed form but you definitely are not allowed to sell to stock image agencies without this.

Removal of icons on the desktop

I was always wondering what's the point of icons on desktop. The idea is really silly - most of the time desktop is covered with opened windows so you cannot access the icons without closing or minimizing windows. Good to see KDE 4 got rid of them :) Perhaps I should switch back to KDE; systems which not allow to remove all icons from desktop always annoy me a bit. The only time I see empty desktop I'd rather see an abscured background there.

Same story here. We run puppet with more then 400 servers and it is pain to get it working. There are many issues which show up here and there: memory leaks in both client and server (aka puppetmaster), periodic lock ups and even file corruption. Besides it is quite slow. I'd migrate us to something better but not sure what to switch to. Besides we've invested into puppet based infrastructure too much by now to make the switch easy. One problem with puppet is that it looks very nice at first when you have small infrastructure with small number of servers and simple manifests. As your infrastructure grows you hit all scalability issues with puppet but by this time switching from it is hard.

For anyone interested into how to get puppet work for any real work load this is what we do:

• Run Puppet under Apache+Mongrel. By default it runs using WEBrick what breaks easily under any moderate load. Plus using Apache you can run multiple backends what helps if you have multi-core server for puppetmaster. Alternatively you can use Nginx+Mongrel or other web server with proxying capabilities + Mongrel.
• Restart backends from time to time because they leak memory. We have a cron job to do this every 15 minutes (yes, it is that bad).
• Puppetmaster has a cache which we saw to get corrupted sometimes. Our "fix" is to delete it before each restart.
• Do not run puppet client as daemon. Run it as a cron job. Puppet client when run as daemon leaks memory and get stuck from time to time. In our cron job we add random sleep before starting client to make sure requests do not hit server at the same time.
• Never serve big files over puppet (i.e. from its filestore). Puppet does a number of stupid things with big files like say reading them into memory first before serving it to puppet client. If you need to distribute big files use other means (HTTP, FTP, NFS, etc).