I don’t want to lie. Being me has considerable advantages. But it has some  challenges, too – like the way I learn Greek:

The Challenge

1) Sit down to learn Biblical Greek.
2)Wow – Mounce’s BBG has a lot of vocab!
3)Cool! Mounce has audio files for every single vocab word!
4)I should listen to Mounce say all the vocab so I can learn to pronounce them right – it will be easier to do this now, so I don’t make a mental association with incorrect pronunciations…
5)But the vocab is spread throughout all the lessons and it’s so slow clicking through all the web pages…
6)This will surely go quicker if I do things The Right Way…

Finding The Right Way

7)The files are all Quicktime…
8)Install Quicktime.
9)I should cache the audio-files locally… and I can maybe even convert them into .mp3 so I can play them in a playlist!
8)Open terminal ==> ‘man wget‘
9)Hmm… But I need to get the file locations from the links! Is this a job for regex?
10)Save the source for thirty web-pages for link extraction.
11)Oh no! The .html files only link to the individual html pages for each word – there are about 300 of them! Time to ramp it up… oh, IT’S ON.

Getting Ready for Battle

12)Is it finally time to learn BASH scripting or should I brush up on my Ruby? Let’s go with Ruby.
13)Which Ruby screen scraping library should I use to handle the html pages? (Google them) Scrubyt isn’t actively developed, should I use Nokogiri or Anemone? I think Nokogiri…
14) Oh wait… Ruby isn’t installed – what’s the current best practice for install on Ubuntu? It seems to be the Ruby Version Manager, but the Ubuntu packages don’t seem to work right. I should probably install RubyGems manually…
15)RVM install botches horribly.
16)Ok, we’ll just use the old ruby1.8 with Nokogiri then…
17)But I don’t have a good editor installed and I haven’t used one in about a year – I wonder which is top of the hill? (Research Netbeans & Eclipse) Hmm… they’re chunky! (Install Geany and Redcar and jruby)
18)Hey, I’m almost ready to start playing around with editors and finding which one I like better… And after this I can figure out how to use Nokogiri! Isn’t Biblical Greek fun?

Battle is Joined

19)Spend three hours writing program that uses nokogiri
20)Finally! Downloading the .mov files…
21)Discover that running wget through Ruby means the downloaded files are incomplete
22)Change the program to generate command line scripts
23)Think about manually running 30 command line scripts. Oh Goody! I get to brush up on bash scripting after all.
24)Google “Parameter Extensions” and remind myself how a bash “for loop” works.
25)End up using ‘find’ to concatenate all the files into a single script.
26)Download the .mov files.

Cleaning up the Battlefield

27) No, really, that’s about it… All that’s needed now is to write my bash scripts to convert all the .mov to .mp3 (after installing the proper codecs, of course) and to generate the .m3u playlist files so I can play all the vocabulary for a given chapter.
28)But I guess once that’s done, I can fire up Audacity and record myself reciting all the words after Mounce – this way I can make my own .mp3s with my own voice, using correct pronunciation *and* Fwith the definitions – so it will make it easier to memorize! Also I can share those
29)But I don’t know about that… that might be a bit of a distraction from the task at hand…

Isn’t learning Biblical Greek Fun?

Addendum

30)As with anything in our modern scientific world, if you didn’t write it down, it didn’t happen… Laboratory rule #1. The last step is, obviously, to write the blog post!

UPDATE: It turns out that downloading via manual wget scripts or even single wget commands clips off the ends of some of the QuickTime files. Weird. The same thing happens if you update your Nokogiri scraping program to generate ‘curl’ download scripts as well. Even if you open them in Firefox and ‘Save As’ you will get clipped versions of the files

But they play all right on the website… Maybe I’ll just use that

This is why I hate computers.

With the rise of the social internet, we’re seeing less privacy and more privacy through obscurity.

With regular, good old fashion privacy, people don’t have information about you.

With privacy through obscurity, however, people have  the information – they just don’t know it. It’s not really private, but it’s obscure enough to give the impression of privacy.

I just don’t have time to read all the feeds of my 200 friends on Facebook. Are their status updates obscure? Yes. Are they private? No.

This leads to the dangerous situation where my life feels like it is private (because people aren’t actually responding to my feed) when in reality it isn’t.

The danger is that obscurity is easy to destroy (it can be as simple as reminding someone that someone else exists).

It can be broken in subtle ways, such as facebook automatically running facial recognition on your photographs and suggesting people you can tag.

Do I trust facebook to tell me who I should be thinking about?

The real short version is this:

There are people online. Some of them are interested in what you are interested in. If you are respectful, they will let you talk to them.

If you want to market online you have to understand how things work online. Now, the secret to how thing work online is that they work exactly the same as anywhere else.

People clump into different communities around commonalities.

In the real world, communities are mostly built around geographic commonality. Because that doesn’t mean much online, online commonalities are generally built around interests.

What do online communities look like? They look like forums. They look like blogs where the same group of people comment and discuss with each other over an extended period of time. They look like rings of blogs. Often they mirror pre-existing communities.

But generally they focus on points of interest. Knitting. Art. Geek Stuff. Theology.

Because the internet is so vast, the chances are that if you name a topic there’s a community that’s interested in discussing it. And because the internet is such a wonderful place, you generally have a magic microphone that lets you speak to every single member of that community at once.

Of course, with great power comes great responsibility, and if you are a moron with your massive microphone, you become a massive moron. The fact that it’s the internet doesn’t make the people in a community any less people.

Introduce yourself to people respectfully. Value them. Show that you value them. Be genuine.

It’s not acceptable to use people or to try and get to know them just for the purposes of advertising. Don’t be fake. Unsolicited advertising is spamming.

What is acceptable is to offer them something of value and to respect their decisions through asking them if it’s okay for you to tell them about something. This isn’t any different than real life. Except… That if you show that you are respectful and compassionate and interested in exchanging something of value with them, you have at your disposal a pretty sweet mic.

From the BBC Article:

The idea would enable a given app to work, for example, on a web-ready television, in a car and on a mobile, no matter the makers of the devices.

However, industry insiders say the idea is unlikely to get off the ground.

Yeah, no kidding. Ima shut up now.  Read the koan.

It was written clearly, with no unnecessary words or phrases. Every bit of the text was meaningful. Nothing was repeated. No definitions were duplicated.

Information was managed in multiple documents with clear linking between documents. Referential Integrity was maintained – when I looked up Division 5.1 in the regulation, it was actually there.

Yes, it was complicated. But I really don’t see how it could have been made any less complicated than it was. There wasn’t really any fat to it.

In short, it was a masterful example of how to manage information in large, complicated documents. Hooray for lawyers!

The other people who are worried about management of information in large, complicated documents are, of course, computer programmers.

On a daily basis, programmers suffer the headaches of redundant or conflicting definitions, poor and confusing organization, and obscure or complex language.

To a certain degree, what the large body of text says is irrelevant. So long as the text is:

• Meaningful
• Subject to demands of rigour and correctness
• Complex and intradependent

then the text is subject to certain principles of organization.

Don’t repeat things. If you repeat a definition, someone will change it in one place and miss another, and, hey presto, conflicting definitions.

Don’t use more text when less will do. Remove any unnecessary complexity.

Break the text up into labeled sections and subsections. Represent dependencies and links between sections and subsections in a clear, unambiguous fashion.

Is this similarity of text an accident? Is there a reason laws are called “Codes?”

The body of law is, in a sense, an executable program. It is taken, along with some circumstantial inputs, and given to an executor which evaluates the inputs according to the law, and produces a result.

It’s an incredibly complex program, because its processor, unlike a computer, is not deterministic. Give a modern computer processor the same inputs twice and it will give you the same result twice. (That’s kind of an important property of modern CPUs – it would a bad thing if they were inventive when adding numbers).

Legal processors, for all the effort expended to ensure a consistent judiciary, are non deterministic. They are human. This is largely a good thing, because humans turn out to be far more flexible, adaptable and fault tolerant than computers.

And lawyers, it seems, do a far better job of writing code than programmers. No wonder they get paid the big bucks.

It turns out that if you’re clever with a webpage, you can pretend to be a visitor’s PC for the purpose of finding out the ID number of their router.

Well, that doesn’t sound so bad. A little technical, perhaps, but not so bad.

Umm… unless Google happened to drive down everyone’s street and take notes on exactly where everyone’s router was. And made that information available as a service. So if you *really* want to stalk someone (well, stalk their router, which is often good enough) now you can! Just get them to visit your tricky website and you know where they live!

Whenever a big accident of some kind happens, like a plane crash or an oil spill or something like that, people usually get frustrated because the cause takes a long time to figure out.

The truth is, there are almost always multiple causes, multiple failures which allowed the catastrophe to occur. The pilot was tired and the ground crew slipped up.  And the weather was rocky. And, and, and… Generally security for these things is so layered that multiple failures have to occur before a major disaster happens.

The problem is, that’s an accident. Unintentional.

Security is harder than accident prevention, because in security, there is an intelligent, malicious attacker who is actively trying to combine systems in the worst possible way.

The offshoot of this is that even if one system is arguably secure, it can still participate in catastrophic failure if one of the systems it interacts with is compromised.  You cannot make it secure – you have to make all the possible combinations of systems secure.

Which is why producing systems of jaw-dropping power is a bad idea, unless you can show jaw-droppingly proportionate benefit.

I guess I just wish that someone at Google had stopped saying, “This is so cool” long enough to ask, “Is this really a good idea?”

“Deploying coherent packages of software functionality as loosely coupled, coarse-grained services delivers dramatically improved application flexibility, allowing enterprises to continuously adapt constellations of services to keep IT capabilities aligned with business goals.”

If I ever want to “continuously adapt constellations of services”, I’ll let you know.

Recently Facebook “updated” its interface. I noticed no great improvement, but the bugs really riled me. Try to write a comment… doesn’t work. Okay then.

When I logged in to my yahoo mail, there was a little box asking me, “What are you doing right now?”

Do you remember back when e-mail was just e-mail and your webmail provider was happy just to show you ads and track your browsing with cookies? When it didn’t feel the need to pry into your current goings on? Why does my webmail care what I’m doing right now?!?

I’m trying to check my e-mail. Grrr.

Oh look… In my inbox is an e-mail from Amazon telling me I can save 14%… on a book I already bought through them. In case… I want a second copy? If you’re going to run a smart recommendation system, at least make it smart.

Google Buzz is already getting deserved flak for:

1. Auto-populating your followers lists with the people you email the most
2. Making that list public by default.

Honestly, who thought it was a good idea to DISCLOSE TO THE WHOLE WIDE WORLD THE LIST OF WHO YOU TALK TO MOST WITHOUT EVER ASKING YOU???

When all you ever wanted from Google’s webmail service was, umm.. WEBMAIL!!!

Whatever happened to not bait and switching your clients?

I get that these are all free services. I get that they owe me nothing. But all I see right now is a bunch of big companies who see me as a tiny little chip in their battle to exert their control over the web. I get that everything they’re doing is probably allowed by the Terms of Service that I agreed to without reading carefully.

I get that I’m just angry and cantankerous because of other things and that they’re just receiving the brunt of it.

But they’re still abusing me. With broken software. And it hurts.

Pain, ladies and gentleman. Web 2.0 pain.

But it’s still better than AOL.

Oh, AOL. All of the sleaze, but none of the rounded corners and unoffensive light blue color schemes. A walled garden community where customers were a valuable resource to be exploited.

You were truly visionary, but you were 15 years ahead of your time.

Thank God for Lent. I’m not sure how I’d survive computers, otherwise.