Monthly Archives: April 2006

Rusty on floating point (and keeping neat code)

Posted on by
1

Rusty talks about the “fun” of floating point and how this all ties into Wesnoth.

Platform consistency is certainly a good thing – so I’m guessing the attack_prediction code isn’t run by each node in a network game in a way where machines could disagree on the outcome.

This does however bring up an interesting thing. What if, in the future, it was going to be on a per-node basis and people wanted it to be consistent. How do you warn that this isn’t the case (to somebody who is really just reading the docs on this function)?

Is it easy (or is there even a good way) to separate code that’s on one machine versus every one? In NDB we have some protocols where some things are done on a master and others on the slaves (and sometimes, when we go back to refactor the code, we move some of this stuff around – e.g. some work on the BACKUP block that I did a while ago).

In NDB we rely on separate documentation (a diagram showing what signals go where and from who) and keep the code for executing the signals together in the code. We require the coder to think when they’re changing things about where the code is going to be executed (on the master, the slave or both).

We’ve also started to get some better habits in naming structures that are only going to be filled out on the master (or slave) or both. Writing code that looks at the wrong thing has been a source of bugs (especially while hacking on something) that are annoying to track down.

So how do we have these functions that in some cases shouldn’t be used (e.g. when consistency across platforms is important, or should only be used on the slave side of a distributed protocol)? Or rather, how do we warn others (and ourselves) from getting it wrong in the future?

Is the ultimate answer just that “you should read the code and understand it before you use it”? Probably, because any comments are going to be out of date anyway….

i now look forward to some sort of discussion.

Beat on “state of the dolphin” (or: Why Software is never really ready until a .20 release)

Posted on by
6

Beat Vontobel blogs about “fuþark: The silence of futhark and the state of the dolphin” which is basically about how he’s found that the 5.0.20 release of MySQL is when the 5.0 release is really starting to shine.

This confirms my theory (that I’ve had for quite a while now… like years) that a software release is never really mature until it hits about .20 (that’s dot twenty, not dot two).

When something reaches .10 (dot ten) it’s no longer going to be annoying for most uses, but .20 means that you’re going to be happy. Don’t ask me really why this is the case, but it is.

Think about the 2.6 kernel (yes, Linux Kernel – honestly, you think i was talking about something else?). At about 2.6.10, it would no longer be a pain to use and get things going – everything was starting to be smooth. As we’re getting closer to .20, things are getting better too. Mind you, everything here does run 2.6 now (and so does my mum’s machine – which is always a good sign of something being ready). With 2.4 hitting .20 – you’d never even think about using 2.2, 2.4 was perfect (except when you wanted 2.6).

GNOME (and everything attached to it) is getting to be a really good desktop – ever since about the 2.10 release I’ve been using just much more of the GNOMEy way of doing things because they’re actually getting useful and usable (don’t get me wrong, previous releases were good too – but a lot more things annoyed me). As the releases have progressed, I’m increasingly convinced that 2.20 will be the “we’re here” release. 2.14 is a lot better, but there’s still a bunch of stuff that has to be done before it’s totally kick-ass.

There are no surprises in MySQL 4.0 (it’s past .20 – at .26 now). Everybody knows and trusts it. 4.1 is at 4.1.18 – which is about as good as a .20 and it’s a pretty happy release. But due to 4.0 being rather solid – a lot of people have just stuck there. We’re seeing a bunch move to 5.0 – but my theory is that this will be 5.0.20 or above. Hrrm… anybody see a pattern?

MySQL 5.1 is at 5.1.10 (or so) and it’s stopped being annoying, and that great march towards a .20 is healthy and active.

GCC 2.95 had a lot of respect for a very long time (now it’s just a bit old). Note that .95 is higher than .20 :)

EMACS is at version 21, but ed is only at .2 (hrrm.. and which is used by more people as their editor i wonder).

aptitude at 0.2.15 (getting to .20) – while apt is at 0.6.40 (above .20). RPM is only at 4.0.4 – so a bit to go there :)

The version of postgresql is 7.5.9 over here… so getting to the .1 stage, but away from the .20. (now I’m going to watch comments fill up with postgesql guys going on about something, i just know it :) But there is 7.3.14 – a lot closer to .20!

MythTV is at 0.19 – getting closer to the .20 release (it’s a lot better than even just a few releases ago).

(versions here mostly taken from whatever ubuntu 5.04 has)

Note that attempting to skip a whole bunch of versions and label your software 95, 98, 2003 or whatever doesn’t get you “.20” status. Neither does just skipping to “.20” automatically. It’s about hard work and removing annoying things (we tend to call them bugs).

This is a really stupid metric of software maturity. It is, however, disturbingly accurate.

My MySQL UC2006 talk – more working on it

Posted on by
Reply

MySQL UC 2006 – April 24-27, 2006 – Santa Clara, CA – MySQL Cluster: New Features and Enhancements

So I’ve done some more run throughs to get things running smoother (and made some more edits along the way). At some point I will stop fiddling with the darn thing. It’s going to be fine (repeat, take 8 times and call me in the morning if pain persists).

Some features take a lot longer to explain than others. It’s quite interesting really.

I’ve tried to strike a balance between good overviews of technology and the down-and-dirty details. Hopefully it’s a good balance and fits the audience. I’m assuming a bit of knowledge about Cluster (I think the SQL knowledge should be a given – it’s a technical talk at the MySQL UC!). I’m expecting to be thrown a bunch of questions througouht – and hopefully really good questions (and one’s that I’ve anticipated and if people only wait for the next thing i was going to say…).

I’ve put a lot of prep into this talk – hopefully the effort shows (in a good way!).

So come along and hear me talk about what we’ve been up to in the lovely land of Cluster.

I love the smell of fresh Basil

Posted on by
Reply

even after you wash your hands, the smell is still there – it’s great! Of course, it’s bad if you’re in public and keep smelling your hands – but Basil does smell really good.

I shall now nervously look around the room to see if anybody is going to see me smell that fresh Basil smell that’s permeated my hands.

I made wikid past sauce tonight.

Perth Penguinista is back

Posted on by
Reply

Perth Penguinista: How to trash a non-life

It’s really great to see you back blogging Leon.  Here too (Perth Penguinista: Free in many ways). I have a few friends who work (or have worked) in hospitals. Let us just say that the tales of computer woes are not to be underestimated.
So cheers to the hospital staff that’s made sure the words still flow (and make sense!)

Incidently, I found the DVDs of the first two series of Scrubs for a reasonable price the other day, So for the past few days I’ve been watching a lot of Scrubs.

final prep on UC presentation (and new toy)

Posted on by
2

I bought a new toy yesterday (and about time I did). A Logitech presentation clicker thingy:

Picture(14) (1).jpg

It has the laser pointer, the forward and backwards slide buttons and, arguably most interestingly, a built-in timer with vibrate alert.

What’s annoying is that the forward/back is done by page up and page down – and this doesn’t work for the “appear on click” thing for OO.org. Luckily for me, I just about never use that “feature” as the in version of OOo that Ubuntu ships in their stable release (5.04) is just too darn buggy in that area. I do sometimes wonder if people use the stable release of their product for any real work.

But it’s a nice little device and seems to be an improvement of the using the remote control feature of my phone to do the same thing (if you didn’t do anything for X seconds, it disconnected and had to renegotiate something that took a few seconds).

RealNetworks is still clueless (and amazingly not broke yet)

Posted on by
1

(note that this entry is in the “Inciting Hatred” category. it is not meant to be well argued or anything. I’m ranting. get-over-it).

RealNetworks rep to Linux: DRM or die!

You need DRM about as much as you need anything containing the word “rectal”. Nothing good is prefixed by the word “rectal”. Even if you like it that way – the only time the word rectal is ever used in that context is to do with disease.

Mind you, if you consider all the shit music that’s pumped out these days, maybe if it’s all locked up in DRM and I never see or hear it the world will be a better place. That’s right Robbie Williams, I’m looking at you.

Can anybody remember a time when Real shipped a product that anybody actually wanted to use? No? Well, that’s because they never have. Apart from the wow factor of being able to get audio (and then video) over the Interweb (err… net it was back then)  it just felt like they hated you. I am still bitter about that linux ppc build that couldn’t even pause reliably.

Even Microsoft did better with their Windows Masterba…err.. Media Player. For a start, I’ve seen somebody use it in the past six years.

So an irrelevant company is saying something stupid and wants to rob me of my freedom.
Say NO to DRM. Also say no to drugs – but mostly say NO to DRM.

Moo, You Bloody Choir

Posted on by
Reply

Augie March – Moo, You Bloody Choir

While I was away (I think I was in Stockholm at the time… not sure actually), Augie March were doing signings of their new album at JB in the city. Jessie volunteered to go get me a copy. So when I got home, I had this waiting for me:
Moo, You Bloody Choir

with it signed by the band:

Moo, You Bloody Choir (signed)

It’s an awesome album too. I’ve really liked One Crowded Hour since I first heard it (a long time ago now) and am really pleased that it got onto a release. I’m also a fan of Clockwork and Vernoona and a whole bunch of other tracks.

They’ve got some gigs coming up, hoping I’ll be able to get to them.

doxygen loves the RAM

Posted on by
2

Why when running doxygen over the mysql tree (5.0 or 5.1) do I have a process with 590MB of RSS memory?

Not exactly inspiring confidence. Although I guess I’m lucky because I have the RAM to do that in (on any box around here I actually use frequently).

The output of doxygen can be really useful when trying to learn (or remember) the relationships between various bits of code. I find it a bit faster than switching between buffers in an editor and then trying to remember where some class was defined. links are a good thing.

It’d be great if we switched all our public API docs to doxygen, as the output really is quite nice. In fact, internal APIs wouldn’t be bad either. Although, naturally, the real documentation is the source, which (luckily) the doxygen output also makes easy to view.

I’ve rigged up this script to automatically pull the latest out of the public repository (using the free bk client) and generate doxygen docs. About time I share this with the world. get_trees.sh you also need the doxygen template (rename it to Doxyfile.template in the same directory as get_trees.sh)

I run this in cron @daily.

my phpbms branch

Posted on by
2

I’ve had to fix a few small bugs in the release of phpbms. So I’ve put my bzr archives up

bzr clone http://www.flamingspork.com/src/bms.upstream/

and

bzr clone http://www.flamingspork.com/src/bms.stewart/

Hopefully there’ll be another release soon that incorporates these fixes – some are on the sourceforge page and some are in the source repo.

totally quivering over phpBMS

Posted on by
2

phpBMS

Basically I want something to generate invoices for me. This should greatly help in a bunch of things – namely not being a retard and fucking it up every month.

Primarily I want to just be able to *not* have a whole bunch of spreadsheet files (one for each month of work plus one for each months expenses) and actually have something that works and takes a lot of the pain away for me.

Then I can do queries to fill out stuff for the tax office.

I think phpBMS fufills this for me. In fact, I’m very much inclined to migrate to it right now.

It stores all its data in a MySQL Database (which is nice, as I use that – and like it). It also means I can do arbitrary queries (in fact, the queries it does are viewable via the Web UI – funky!)

It’s even buzzword compliant with AJAX.