Dear Journalists: Bits and Bytes

Dear Journalist type person,

There's something we must discuss. You see, you've been making a very basic mistake in many of your articles when it comes to writing about the Internet, and specifically Internet speeds. Let's take a look at a small quote:

...unless you have an internet connection of impossible speeds. (Mine is nominally 10MB, by the way, which in practice means maximum download speeds of 1.4 megabytes per second).

(source: Rock-Paper-Shotgun).

Can you spot the problem here? Internet speeds are measured in megabits per second. The symbol for 'bits' is a lower-case 'b', so an Internet connection that's 10 Megabits per second could be written as "10 Mbps". I guess if you're feeling lazy you could leave off the "ps", and end up with "10 Mb" (although it's a really sloppy thing to do), but NEVER "10 MB" - that means something else entirely.

Modern PCs use bytes that contain 8 bits. The correct symbol for a byte is an upper-case 'B'. so "10 MB" means "ten mega-bytes", not mega-bits, which is probably what you meant when you were describing the speed of your Internet connection.

Back to our Internet connection that runs at 10Mbps. It's unfortunate that speeds are measured in bits, because a much more useful measure is bytes per second, since that's how we deal with data sizes. We know that a CD ISO image is likely to be around 700 MB, an MP3 file around 3 MB, and an image from a digital camera to be around 1 MB. To convert our 10Mbps connection speed to megabytes per second, we divide by 8, and get 1.25MBps. However, this is the theoretical maximum speed, and there's a lot of overhead in any network connection, so in practise it's unlikely you will experience anything close to this maximum speed.

If your eyes glazed over, or perheps you felt light-headed reading that, here are a few take-home points to make it easier for you:

• Connection speeds are measured in megabits-per-second. The correct unit symbol for this is "Mbps".
• Files are measured in Megabytes.
• A Byte has 8 bits. So to turn your connection speed into something useful, divide the number by 8 and make the unit symbol "MBps".

I would be honoured if you'd consider this small point next time you go to write online. Some of us are acutely sensitive to these matters, and you really don't want to upset the geeks of this world.

Kind Regards,

KiwiPyCon 2012 will be in Dunedin

It is a great honour and privilege to be involved in bringing the New Zealand Python developers conference (KiwiPyCon) to Dunedin in 2012. I intend to blog regularly with details of next year's conference as they emerge, so watch this space for more information over time!

Finally, a huge "Thank You" to Tim McNamara for organising the best KiwiPyCon to date. I have a huge task ahead of me, and he's raised the bar very high indeed.

Have any good ideas for next year? Please let us know!

Sloecode 1.1 - get your bugreports in now!

I'm at KiwiPyCon in Wellington, New Zealand. It turns out, being surrounded by geeks is great inspiration & motivation. Because of this, I've created a Sloecode 1.1 milestone in launchpad. If you have any bugreports or feature reuests, now's the time to get them in!

...I can't believe I'm asking people for bug-reports. This feels wrong...

Sloecode: now with 100% more website

Sloecode now has a real website! It's a rather minimalist affair at present, but it's better than nothing!

Since I last blogged, the setup instructions for the Sloecode server have changed a lot. The new instructions are available on the new site.

Representin' the home-boys!

That's *exactly* the kind of blog title that's liable to get me in trouble down the road. Faux street swagger doesn't age well.

In other news, I'm presenting a talk at this years NZ Python conference KiwiPyCon. Their website does not have the schedule visible yet, but there is a list of talks available.

I'm looking forward to meeting other python developers, and maybe getting some help with Sloecode!

Unity Workarounds

I'm currently running the unity desktop shell on my development computer. I got fed up with KDE's lack of support, and the gnome-shell just annoyed me. Unity seems like a good start, but suffers from several problems. I've found workarounds for all of these, and thought I'd post them below.

Scrollbar Corruption:


One of the major changes in unity is the thin scrollbars (technically called "overlay scrollbars", I believe), which are enabled by default for all GTK applications. This is what I'm talking about (image from techgarage.com):

These work fine for native GTK applications, but applications built using the wxPython toolkit suffer from render corruption when you try and scroll a styled text control (bug report). This bug manifests itself in many applications, including my current favorite development text editor editra. Here's a screenshot of the issue:

Solution:


The problem can be avoided by turning the thin scrollbars off. How do you do that? Easy - launch the application with the "LIBOVERLAY_SCROLLBAR" environment variable set to 0, like this:

You can of course edit the menu item for editra to include this, or add the line "export LIBOVERLAY_SCROLLBAR=0" to your ~/.xprofile file (create it if it doesn't exist).

Panel Corruption:


Whenever I change a setting in the compiz config settings manager (ccsm), the top panel of my unity desktop becomes corrupted. Previously the only way to fix this was to log out and log back in again (effectively restarting unity). Here's a screenshot of the problem:

Solution:


When this happens you can restart unity without having to logout, or close any applications. Simply run (either by pressing Alt+F2, or in a terminal):

unity --replace

CAPTCHA Fail

I can understand the need to filter humans from machines on the Internet, but we need to do a much better job. It seems like the most common captcha software I run into these days is reCAPTCHA. The system has some noble goals - specifically, users who submit captchas are actually helping digitize books. However, here's the problem:

What the hell does that say? The first word is obviously "the", the second? "Claccupr?". Perhaps I have bad eyesight, but I often get these wrong. Here's an alternative proposal:

Instead of asking humans to perform text recognition (something computers can already do reasonably well), ask them to do something that comes naturally to humans, but not to computers. How about asking the user to pick which image they might find at a beach. Or ask the user to pick the image that represents the largest object in real life. Use context that humans understand instinctively, but computers do not.

Any ideas? Someone should build it!