Bootstrap Diaries: Knowledge is cheap.

Very cheap in fact.  Obviously there’s the Google/Bing/(Name your other favourite search engine here).

When it comes to knowledge for Datasentiment I had a bit of a head start from 2002 when I worked for a data mining company. I learned an awful lot about things not to do to your employees, how bad some Phd coder’s Java can be and watching reams and reams of data from a well know supermarket that couldn’t compete with the clubcard.

For those who met me last year when I was talking about free tables the like on the phone, well the next three months were an unfocused blur and the months between October and March have been the most rewarding, productive and exciting times.

Any gaps in knowledge were basically filled with the books below:

One of those books was listed on Amazon at £0.00, yup it was in their warehouse and they wanted rid of it, all I had to do was pay the postage.  The majority of these books were below five pence and all of them have domain knowledge that I couldn’t easily find on Google.  Tesco certainly don’t give their clubcard secrets away.

There’s no excuse for “I don’t know” because i) The knowledge is out there and ii) there’s probably a dozen people who’ve asked the same question.

Post-It notes came in handy for my copy of Rework when it landed.

The Bootstrap Diaries – Part 1.

There are just days when you have to go it alone.  For those who don’t know I’m bootstrapping a company from scratch.  I thought it might be a good idea to chip in some blog posts along the now that I can talk about some of these things.

Since my last job finished at the end of June I’ve been wondering the best way to take things forward.  Ask anyone who really knows me and I can have a list of ideas that are ready to be coded.  The most interesting thing for me is that this is first start up where I’ve been asking a lot of opinion of the product I’m doing (it’s still in stealth, there’s only a chosen few who know).  

So here we go, the bold pointed things to keep in mind.

Writing it down on a one pager – It’s vitally important to write down on one page what your product is going to do, who the target market(s) are and what the total size of the market is.  From there you’ve got focus and what you think it’s all worth.  Ignore this step at your peril.

Blag, steal, borrow and blag again – If you design, get a good back end coder.  If you code then get a designer.  This also goes for business advice.  Some of the people I hold in high regard are willing to chip in and keep me on the right path.  Finally, don’t forget a polite thank you goes a long way.  

Crap code is fine – Product first, coding refactoring later.  There’s no point going over the finer detail of SCRUM or Agile methods if it’s not generating revenue. 

Market share or revenue? – In the initial stages it’s all about generating revenue (especially if you are a service company).  If you’re hellbent on creating a brand or a world changing-everyone-has-to-have-this iPhone app then market share is your aim.

What time do you realistically have? – If you are already working then it’s extra hours. I know some that are up at 5am and work through ’til 9am. I know some who do the late shift.  There are times I do both and within time I suffer for it.  Also, if you have family they will only put up with your Branson like enthusiasm for so long.

Learn to do a basic cash flow forecast – Yeah it’s boring but I bet if I asked you now you wouldn’t have a notion what your costs are for the next 12 months.  A simple spreadsheet is fine.  There are a loads of good cashflow templates out there on the internet.

Get an advisory board of people you trust – Ultra important this one.  I have three people that I bounce stuff off in various parts of the world.  I also have a number of industry contacts who I bounce ideas off.  I have one friend who I can just cyberly puke on when I’m having a bad day (and they with me if needed).  

Don’t over do the networking while you are developing the idea – Any networking event the obvious question is, “what do you do?”.  When you are just about to beta test fair enough.  From day dot without a single jot of code, forget it.  You could be using your time on better things.  One thing I’ve noticed over the last five months is that there are an awful lot of people who can talk the talk but when you really press them…. you pretty much know it will come to nothing.

Some of the above I actually covered in my talk at Barcamp Derry in October but they still ring true whatever the time of year.

The new Java Store and the main core of the problem.

Three cheers for the new Java Store (though I can’t see as I’m in Northern Ireland, can’t sign up, can’t join, well not for the time being). From reading press releases and the rest it sounds good on paper but I’m far from convinced that Sun can actually deliver.

Eric Klein said,“We’re building new business opportunities and are thrilled to launch the Java Store to connect developers with hundreds of millions of Java users. The Java Store will become the destination for the most interesting, useful and entertaining Java software and content.”

And here starts the main point of the issue. Not everyone with a Java runtime knows they have a Java runtime. It’s all very well to say there’s 2.4M mobile devices but out of those how many truely know they have Java on the phone and they can download stuff. Not 2.4 million for sure. Set top boxes are another good example, I don’t know many who can download software for them, so it’s all very well saying there’s 800M of them.

At present the Java Store is a big marketing exercise from what I can see. It’s a good idea and I desperately want to believe it will work. Everyone is looking at Nokia Ovi and a bit of a tramsmash launch. I’d hate to see the Java Store go the same way. We’re going to end up with an app store deluge and the single one fact remains, Apple is doing it better than anyone else regardless of whether you like Apple or not.

Sun going into the Java App Store business.

Mainly to promote the Java runtime but it makes sense. Jonathan Schwartz’ blog from Monday outlined the new Project Vector in part but he’s leaving the meat on the bone until JavaOne.

And that’s what Project Vector is designed to deliver – Vector is a network service to connect companies of all sizes and types to the roughly one billion Java users all over the world. Vector (which we’ll likely rename the Java Store), has the potential to deliver the world’s largest audience to developers and businesses leveraging Java and JavaFX. What kinds of companies might be interested?

Applications are on application and will have to be real good to get on the space. As with Apple’s way of doing things the apps that will really make it are the ones that get profiled on their site. You need money to make money.

It’s an interesting concept (though you’ll see by the comments that there a few confused about how it’s really going to work for developers). No real mention on J2ME apps either, this looks like a big push for JavaFX on the desktop to get away from all those nasty browers.

Google App Engine: on Eclipse it’s a dream.

My first proper morning of having a look at the Java support for Google App Engine. A tip for anyone who is developing, make sure you have Eclipse and the AppEngine plugins, it works wonders.

Coding the app up is no real problem. If you are used to JDO queries then you’ll be fine, I’m so used to JDBC hell that I had to read the docs. Where the plugins really shine is on deployment. It totally removes the whole command line procedure. Click once, enter your username and password and up it goes (fortunately not in smoke). All done and working.

Not a design classic but here’s the first app.

Java in the cloud: First look at Google App Engine for Java

I was lucky to be one of the first 10,000 developers to get a closer look at the new Google App Engine for Java. Up until now it’s been Python developers who had all the fun.

What’s it got:
The Java environment provides a Java 6 JVM, a Java Servlets interface, and support for standard interfaces to the App Engine scalable datastore and services, such as JDO, JPA, JavaMail, and JCache.

First things first, if you use Eclipse there are some plugins that Google have developed for the development and deployment of Java apps for AppEngine. There’s also a good guide on getting started here.

The problem with these sorts of things is thinking of something to code up. So I’m going to a think and come up with something.

More later…..

The great IDE clutter

Perhaps it’s old age creeping in, perhaps it’s just wanting a simpler life (now a reality since I ditched Bebo, my Twitter account will more than likely head the same way).

Next on my list, development tools. Just too many of them, but then again there’s not one day when I know what’s coming. For Java I use Eclipse all the time, have been since 2002 (I even said nice things about them in 2003). The guys at Silktide have got me into PHPEd for all things PHP, though I have to say the installation did kill my Apache 2 config and removed any trace of PHP. Ruby on Rails, uhm this is a problem child at the moment. I swing from Textpad to Komodo Edit 5.0 and then back again.

In an ideal world everything would run off Eclipse. PHPEd doesn’t seem to really create much of the way of PHP code for me. I want a class, I should be able to pull a table from MySQL and it create a class for me. Older versions of Eclipse support a PHP plugin but as I’m on Ganymede a few of the plugins don’t seem to be supported yet. I’m just too cutting edge sometimes. As for Ruby on Rails, RadRails was built on Eclipse but was replaced by Aptana’s Studio (also for Eclipse as a plugin but, alas, not Ganymede).

Taking a tip from David Heinemeier Hansson, code generators or more importantly scaffolds are a way forward for PHP. I’m just wondering if there’s any use in modifying the Ruby scripts to generate PHP code as well as Ruby code?

The purge continues.