It has recently struck me that one of SELinux's less obvious problemsis that it is a secondary, backup security mechanism. By this I meanthat no current system depends on SELinux for its primary securitymechanism. Instead the primary security mechanism for Linux is stillUnix permissions and UIDs, and on standard distribution setups SELinuxis there purely to limit the damage if a program has a bug.

(One of the signs of this is that SELinux denials are treated as a bigdeal. Imagine if your Linux system alerted

I think that my core problem with SELinux is that it is very fragile.By this I don't mean that it's buggy and so breaks in normal operation(because as far as I can tell it generally doesn't), but instead thata SELinux configured system is not resilient to changes. If you runthe system exactly as your distribution delivered it, with everythingin the same places and with the same user-ids and so on, the resultgenerally works.

(You start running into

In an entry, Zed Shaw recently wrote :

P.S. I have a long bet that SELinux is an NSA backdoor. Any takers?

I'm an optimist, so I think that there's another good explanation forSELinux and all of the effort that the NSA has poured into it. Mytheory goes like this:

Basically, the government has a problem. Historically and forgood reasons , it is theonly customer for really secure systems (ie, rainbow book level paranoid security).This has

The US government (and to a less public extent other governments) havespent years developing a very robust and theoretically sound model ofcomputer security, complete with rating levels and all sorts of goodthings. They even documented this thoroughly in what is called therainbow series of books .

You might think that a government developed set of computer securitystandards would be widely adopted by industry and broadly used, in thesame way that other government-developed standards generally are. Youwould be wrong; the industry reaction to all of

One thing that creates the the SSL CA problem is that websites have no secure out of bandmechanism for asserting what SSL certificates they use (or that shouldbe accepted from them). With such out of band information, incorrectlyissued SSL certificates wouldn't be a concern because they wouldn'twork.

Well, 'secure' is an imprecise word here. The problem with SSL CAsis that they create a situation where SSL has multiple trust roots;in this situation your security is only as strong as the

When I found my example of careful and clever spammers compromisinga blog , I of course thought about letting theblog's owner know about the problem so they could clean it up. But themore I thought seriously about doing that, the more uncertain I got,because of the fundamental problem here: how do I make sure that I'mnotifying the blog's owner instead of handing my email address to thespammer?

Let's turn this around to the flipside question: if you're a

I consider all of these to be corollaries to my first set of modestsuggestions , or more exactly to the core ideabehind my suggestions, that being that you should arrange it so thatthe open issues in your bug tracker are things that you are fixing, notwishlist items, not things that are maybe going to be fixed some time inthe future. Stuff you are fixing now, or close to now.

This implies that you need a way to explicitly defer issues, in orderto deal with things that you

The Posix standards for Unix sometimes get a bad rap (there are somecrazy things in there, and some not so great ones), especially forthe bits that they invented, but every so often I think that they didsomething absolutely right. The new Posix $(...) syntax for commandsubstitution is one shining example of this; it is superior to theoriginal Bourne shell syntax of `...` in so many ways. Here's someof the reasons that I like it.

I have written grumpy things about the limitations of the CSS overflow property a long time ago , but I've recentlyrealized that it does have one actual use in an environment such as WanderingThoughts . And that is to deal with a problem I periodicallyhave with comments.

The problem with comments on WanderingThoughts is that they can breakthe overall page layout. WT uses a table based layout, so an overly-longline in a comment forces the entire surrounding box to widen to containit. The result creates a table