Why I dislike multi-user

I haven’t talk publicly about this before, but I somewhat dislike multi-user OSes. To me, they just don’t make sense and NO operating system today has a perfect implemenation of it. And even if an OS gets one, the drivers and third party apps will be so badly written that they won’t support the system well. I see that ALL the time on OSX (see: epson drivers), even on Windows doesn’t work great, as for Linux, well, should I go further than the permissions problems on simple things like mounting cameras or having a gazillion personal hidden files on your ~/?

I was not always like that. It all started last Christmas when I needed to create an official legal paper so my brother could do a few things FOR me while I was not in Greece (give him the right to sign stuff for me etc). Well, what I really needed was specific and yet there could be cases where the paper’s explanation was not good enough for what my brother should be able to do and what not. There was no way to have the paper describe exactly the “permissions” I was allowing him to have.

I feel the same way about OSes with multi-user support now. No matter how much you hack it to support many different users, the end-user experience at the end WILL NOT be as good as if the user was the sole administrator of the machine (suppose there are not security conserns). Because no, you don’t want your kid to install a porn client or a beta driver, but you might want your kid to install system-wide the new Firefox version as it’s not that big of a deal. This is a lot like allowing your kid to park your car in the garage because you are currently lazy to do it yourself, but you don’t want your kid to take your car and start painting the town.

What I am trying to say is that there is never a clear cut on permissions. Never! Especially inside a family where there is TRUST for certain things. And so NO SYSTEM in the world, EVER, will be able to fully mirror the intends and purposes and needs of a family PC.

Sure, companies might have admins, but for most things that users would like to be able to do will overwhelm any admin to give persmissions every so often, and for families multi-user just doesn’t work well.

Why do we use multiuser then? Because it’s *good enough*. But it ain’t great and it will never be, unless OSes get artificial intelligence or become part of the family. 🙂

OR, having the PC instant message the administrator on his mobile phone to ask if the user XX should be allowed to do the YY action or not. Overwhelming for the admin, but a good compromise.

Comments are closed as this blog post is now archived.

Lines, paragraphs break automatically. HTML allowed: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>

The URI to TrackBack this blog entry is this. And here is the RSS 2.0 for comments on this post.