Monday, February 19, 2007
But actually what does happen when all space is actually used up? Since I am subscribed to some high volume mailing lists with my mail account from this large internet company it happened to me already sometimes that the account was almost full. To prevent trouble I just deleted some old mails whenever this occurred to happen. Then a colleague told me that what I am doing does not make any sense because he claimed to know that the account can just be filled up further and further, even more than 100%. That sounded strange to me but seemed to be actually an interesting question.
So this is what I did when this happened again for my account last weekend: Nothing. I just waited to see what will happen. Actually he was true in the sense that nothing happened. But not about the results. Actually I suddenly got much less mail to that account. This seemed to be somewhat suspicious to me and I sent a test mail from another mail account to my own mail account from this large internet company. The mail was sent without an error. The mail never reached my mail account from this large internet company. Neither the sender nor the receiver got any error message. The mail was just silently dropped. This is definitely not the error handling you expect from a mail provider.
So, what can we learn from this test:
1. Never trust your colleagues if they tell you some strange and fantastic stories.
2. Never use free mail accounts when you have to rely on receiving your mails.
3. Never expect people to do proper error handling for a case they are proud that it will never be needed. Titanic told this already in history.
Sunday, February 18, 2007
Thursday, February 15, 2007
After all I am not that sure at all that the quality of the most expensive conferences is necessarily the best. Those conferences suffer as well from bad quality assurance during the reviewer process. Currently I have again a paper lying on my desktop where I found two mathematical errors within a very short time. One of those errors is even fundamental for the algorithm thus doing it the way it is described in the paper simply does not give useful results. I wonder why we need a review process if such fundamental errors are not caught there? It seems all the reviewers involved in the process for this papers did not really understand that part of the paper and then just accepted it because they didn't want to confess that they don't understand it.
After all this paper has good contents in general but I would have expected the reviewers either to catch these errors (at least the fundamental one) or to confess that they don't understand the content and opt-out of the review process. --- Ok, I agree that this is another dream of an ideal world...
Friday, February 09, 2007
Wednesday, October 11, 2006
Sunday, September 24, 2006
Monday, August 14, 2006
Anyone with a good suggestion for a _usable_ blogging tool that also allows adding pictures without clicking hundreds of buttons for each?
This and the fact that some features of the crappy web frontend of Blogger did not work at all were the reasons I did not post recently and why I will make all this here less frequent and shorter.
In the meantime I had a meeting with a black bear in the wilderness. It seems the bear was at least as surprised meeting me there as I was meeting him. Thus we both just looked at each other and after some time of mutual inspection we just continued our way.
Yesterday I was hiking this mountain.