Grmbl

So we'll be doing a video recording of a musical in the second weekend of October. Since it worked so well last time, we're going to be doing it with DVswitch. One quite important bit of goodness that we had last time was the fact that we had an intercom; that is, some way for the camera people and the director to talk to eachother. I hadn't set that up; the theatre where the concert was just had that material, we could just use it.

Since the place where we'll be doing this recording does not have such a thing, that meant that in order to repeat the success of last time, we'd be needing something else this time around. So I bought five USB headsets, set up an asterisk installation with conferencing, installed linphone-nox, and tried. Only to find that it didn't work.

Fiddling with the settings didn't help. I got a bit angry, since there's sooo many sound systems in Linux that sometimes it's hard to figure out who's at fault. But no matter, whatever I tried I couldn't get it to work.

After trying for a few hours, I tried using it on the PS3. The packaging did say 'PS3', after all, and perhaps it used some playstation-specific protocol? I'd be surprised if it did, since even their controller uses the standard USB HID protocol, but then it could be. Surprise, surprise, even on the PS3 I couldn't get the microphone to work.

Open up another package, plug in the second headset, and yes, obviously, now the microphone works. Sigh.

Oh well, that's what warranty is for.

Samba

Warning: rant ahead

Someone decided that doing a Windows-specific protocol would be a good idea. Whoever this person is, I hope he rots in hell. But that's not what this post is about.

Someone else decided that doing a free software implementation of that Windows-specific protocol would be a good idea. Hence samba was born. And all was good.

For the next few years, Samba kept implementing more and more features of that particular Windows-specific protocol, which kept getting changed and renamed over the years. Originally, samba could just export files, and you could specify whether users could or could not write to your files. Perhaps it already had passwords at that point—not sure. Later, usernames were added, and the ability to join a Windows domain, and then another type of Windows domain, and eventually a number of other differently deficient difficulties.

As these things were implemented, the ways of configuring samba to properly do all these things was made more and more complex, to the point where it now no longer seems possible to install samba without major repercussions on the 'plain' Linux installation. A bit like having an apache requiring you to add apache-specific usernames through an apache-specific NSS module (winbind anyone?) just so you could export ACLs. Oh, and forget about retaining your existing usernames.

Now I'm not sure whether this is related to a) me having tried winbind at some point in the past, deciding that I didn't like the effects it had, and removing it again (with as a result that it left some traces on the system), b) the system being joined to an ADS domain which might have some special requirements, or c) this particular system having a bit of a weird history that might have fucked up some of the internal state files of samba; but at any rate, it sucks.

It seems to me that somehow, over the years, samba has lost the ability to just function as a minor something that just happens to be one service on a machine—much like apache on my laptop—and now requires one to do a complete overhaul of the entire system. That's just stupid; I am not going to migrate users to their winbind counterparts, thankyouverymuch.

Stupid ********

Of course, though unlikely, it's not impossible that I've missed something. But at any rate, even if that is the case, the hoops one needs to jump through to get to the above situations are far too numerous and complex.

FSF Hardware endorsement criteria

The FSF has published criteria under which it would endorse hardware as being 'free hardware'. My guess on the number of companies that would be interested in this policy in its current form:

0

My reason for that is fairly simple:

Any product-related materials that mention the FSF endorsement must not also carry endorsements or badges related to proprietary software, such as "Works with Windows" or "Made for Mac" badges, because these would give an appearance of legitimacy to those products, and may make users think the product requires them. However, we don't object to clear factual statements informing the user that the product also works with specific proprietary operating systems.

In other words, "if you're going to be using our badge to say that your hardware is free, you must not be using the official badges that show it also works with other operating systems." Or, also: "if you want to make your product more attractive to the 10% of users who prefer free solutions, you must in the process make it less attractive to the 90% of users who don't (yet)."

I don't think that's going to fly.

Bubble maker

I don't believe in bottled water. The reason for that should be obvious to anyone who thinks about it for more than just a second, but in case you can't figure it out:bottled water sucks.

Since I moved out on my own, now over two years ago, I haven't bought the slightest amount of water other than through the tap. This wasn't a conscious decision, originally; it's just that I'm not a fan of non-sparkling water, and that buying 1.5L bottles of sparkling water would quickly reduce to non-sparkling ones by the rate I drink it, and I just didn't think it worth buying for that reason. As I came to realize that tap water is a fine beverage, and that for all practical purposes it is free—or mostly so, compared to bottled water—I stopped even considering the idea to buy bottled water.

But I do still like sparkling water every once in a while; and this weekend, I remembered that about a decade ago, an acquaintance who worked for Philips had shown me a prototype apparatus to generate sparkling water out of plain water in approximately 1L quantities. It was a prototype then, so I could imagine it being a product today. Googling around for a while found no Philips product of that type; but then I did find another company that specializes in such devices. Their devices are fairly cheap (or, at the very least, affordable). And the best part: it doesn't even require electricity (contrary to that prototype device I saw so many years ago).

One is on its way now. Wonder what it'll be like.