I read that antivirus opensource software would be a good idea since the database could be informed and mainained by lots of capable people. However I have also read arguments to the contrary but none really explaining why.
Is an open source antivirus for windows a bad idea or a good idea?
I digress the notion that an open source antivirus database will necessarily be maintained by lots of capable people - this is not how open source works. The principles of open source would entice those that are not motivated by direct financial gain - whether these individuals are highly capable, or not - that is a question of chance or self motivation.
Also, there are two components at play - one is the antivirus engine itself, the second being the definitions database.
While I am certain that a couple of open source antivirus engines are adequate (e.g. ClamAV), maintaining an up-to-date database of virus signatures (and heuristics, for that matter), is a task that requires a vigorous vetting and testing process that would avoid false positives and risk making systems or software inoperable. Not that commercial antivirus vendors are immune to that problem - they are not, but I always look at efforts such as Wikipedia, where it works great to somewhat educate the general public, but it is questionable whether that model would work for an antivirus effort.
The Linux kernel, on the other side, is an open source effort that has had a great success in powering many critical systems. Whether that collaboration model can be used for an antivirus database, where the time from malware discovery to signature creation is critical, remains to be seen.
Overall, I am of the opinion that the idea of "people's" antivirus is good, but the implementation would be a major challenge - the signatures bit especially.
