How to get noticed ...

Right now, I'm the only person who knows about www.friendsbychance.com.  So the next step for me is to get my site noticed. Here's what i'm going to try:

• Email friends
• Use Google webmaster tools
• Send press releases to blogs like TechCrunch
• Comment on lots of blogs and include a link to the site in my signature
  
• Dump some leaflets in letter boxes in my neighbourhood
• Try and make a funny YouTube video (though getting that noticed may be an exercise in itself)

I'll post some results each week.

Blogged with Flock

Friends by Chance

My new web venture is up and running - www.friendsbychance.com. Phew ... at least I've accomplished something entrepreneurial this holidays.

Blogged with Flock

Finishing touches

I'm putting the finishing touches on a project today. Just the words around what the site is, some privacy, and then it's good to go. I'd love to include some sort of blog on the homepage, just as an easy way of keep ing content fresh, but I don't think I'll bother.

I've been reading a few high profile tech blogs recently, and almost all of them have had a complaint about the lack of POP/IMA support from Facebook for mail. I've had idea floating around for this for a while, so I'll try and get a prototype happening as my next project.

Blogged with Flock

Back on deck

Xmas holidays have started, RealJob deadlines were made, and so now I'm back and getting ready to push through a few projects while I have the time off. The plan is to recharge a bit first .... there may not be anything happening until after Xmas.

Blogged with Flock

What have I been doing?

Sifting through mountains of blog posts, mostly. I read an interesting one from Fred Wilson (a VC who blogs) about social networking, and so I decided to email him. His response: "I don’t know of anyone doing exactly this, but I agree with you that it is where social networking is headed". So there you go, I'm on the right track to create the next generation Facebook. Now, all I need is money and time :)

I've also been working on my business plan at PlanHQ. Not for the social networking idea, but for another idea. PlanHQ is pretty cool; it's basically an online business plan application, but one nice side effect is that my plan is effectively "backed-up" off my local PC whenever I work on it. I was trying to complete my plan ASAP, but have realised that the silly season is upon us, so I think it's probably a better idea to approach VC's in the New Year.

Finally, I have a neat little idea for a social-type app that I'm going to do either one weekend or when I have my 2 weeks off over Xmas. I think I'll try out a virtual hosting service like Amazon's EC2 (no link provided ... it's mangled).

Blogged with Flock

Cracked 100

My Facebook app, "At The Moment" has cracked the 100 installed user mark. It's not making any money, but it has been interesting to see the growth over time. I hope at least some people find it interesting ... :)

Blogged with Flock

Biz plan blues

Working on a business plan at the moment. Wow, where to start. I figured the best way to get things done was to put up all the headings, then type away. I changed tack once the headings were up though. What I'm doing is flitting around the headings, brainstorming stuff. Once I have a fair bit down, I'll be able to refine it and see where any holes are. I wish I could just get a VC to read my mind and then give me money ...

Blogged with Flock

On hold

I've decided to put X1 onhold for a week as I try and doc up a fantastic new idea I've had for monetarizing blogs and other web content. Watch this space!!

The plan is to write it up, then just go right for VC money. So that will be the third different approach; the first with MLTA is via my partnership with a friend using SCV, the second is going it completely solo and on the cheap with X1, and now the third will be trying to get funding up front and starting full time.

Blogged with Flock

More progress

Completed a fair amount of X1 on my holiday from RealJob (back tomorrow). I've worked out how to do ReST with Struts2, and also started to sort out the views. Clearspring still looks a winner for me, but it has two types of functional widgets (you can do images but they don't do much). One is web, basic HTML and Javascript, and the other is Flash. Now, I really want to do Flash, but it's over $1k for the license here in Oz. Oh well, HTML it is ...

One other thing I've found is that IE has it's own set of rules. I probably spent a good 2 hrs fixing little IE quirks that work fine in other browsers. I use a Mac (I <3 my Mac) but I use VMWare Fusion (I <3 VMWare Fusion too) to test all my web apps under windows. It certainly beats having to run a separate box.

Blogged with Flock

Clearspring

My new site (codenamed X1) has user content that I want them to be able to share, so I'm investigating Clearspring. So far, I've only read the blurbs, but it seems pretty cool. I'll write about my experiences tomorrow or the next day.

I've had today off from RealJob and have achieved so much. I've basically finished registration and search, while still spending a bit of time with the family. Hopefully one day I'll at least make enough money to support us for a year or two, then I can really go to town. I have probably 5 ideas in the parking lot at the moment, so at least I'll always be busy I guess ...

Blogged with Flock

Almost there

I've been pugging away with X1, and it's almost complete. Most of the back end stuff is there, and some of the front. I've done a few things that can be reused across other projects too, which is always nice.

I've also been working on a methodology for small tech startups, which I have outlined on my consulting website. It's nothing ground breaking, but it does 'separate the wheat from the chaff' so to speak. Many aspects of mainstream methodologies tend to apply only to big business (and even then that's debatable imho), so I've tried to focus on things that really add immediate value to small tech startups.

New Blog

I've started a new blog - The Nu Web. Ok, so names aren't my strong point ...

I'll continue posting here about my start up adventures, but my new blog will be used to post about my web theories (and ideas I won't be pursuing myself). Check it out!!

Musings

I've been thinking a bit about some new projects. I think that there may be a few options for "the next big thing" but I just can't get them straight in my head. I currently have 2 projects on the go, MLTA and X1, but have a couple more ready to start when I get the time. Actually, even X1 is suffering from a lack of commitment on my part :( MLTA is ready to go but waiting for my business partner to move house and start testing and signing up beta participants.

I think I'll start a new blog to capture some thoughts as I go, and keep this one for projects that are happening.

No effort

My Facebook App (At The Moment ...) now has 72 users. It's been growing in drips and drabs without any effort on my part. I am basically using Appsaholic to advertise other apps on my app, then when I've arned a few dollars, spending it advertising my app. It's a slow process but it is no effort and it's growing. Makes me wonder what would happen if I was to sink $100 into advertising it, but I just don't think it's worth the expense, as the revenue per user is very, very low. I think to earn an average income an app would need a very large amount of users who use the app every day. ATM only gets used by a few people per day even though it has over 70 subscribed.

New project

I've done a write up for a new idea I've had, codenamed X1, and passed it on to my business partner for review. Unless he finds anything majorly wrog with it, I'll get cracking on it ASAP.

That's one of the bonuses of having a business partner; the chance to bounce ideas off each other. I often find that a good idea tends to grow better when other people are involved. The different perspective usually shows up areas of opportunity that the original idea didn't cover, as well as showing up any dumb mistakes :)

MLTA in final testing

I've completed all of the coding for MLTA; now I'm just waiting on Zej to do the final testing and to provide some descriptive text for a few areas. I'm currently working on a plan for the next project, but I'm struggling with motivation on week nights due to increased Real Job stress levels. Oh, and Halo 3 has just been released ...

Almost there

The MLTA beta is almost ready ... I've just added the last of the major functionality an all that's left is to add some sensible text in the right places. Oh, and fix any bugs :) Next week I'll be documenting the next idea I think SCV will be working on, as well as adding any fixes to MLTA.

RealJob is in the middle of requirements gathering, and it's getting me down. The traditional waterfall SDLC is so heavyweight and ponderous; it's especially frustrating because the project has already chosen a software package and is really only supposed to be configuring it with some minor customizations. An Agile methodology would absolutely rock this project, but there was never any chance of that happening. It's ironic, because although Agile is seen as risky by most large companies, sticking to the "tried and true" method of the waterfall SDLC can increase the risk of going over budget and failing. They say 9 out of 10 projects fail ... it's probably because they're using the waterfall SDLC and spend 3 months gathering and writing requirements using teams of 20 people.

Progress

Our Facebook app has evened out at around 60 users. I'll leave it up there, but I think it's pretty much done and dusted unless we spend money advertising (which isn't going to happen). There may be the odd growth spike (it jumped 7 users one day) which I'm assuming is someone stumbling upon it and inviting some friends, but I think I'll just chaulk it up as a good research project that's now complete.

MLTA is almost in Beta, I just need to finish a few smaller items. I've found that it's easiest to finish off "boring" bits of apps on the weekends while my daughter naps. I'm fresher than during the week and I can still finish decent sized chunks because she naps for over 2 hours. Weeknights are now reserved for new projects, Xbox360, movies, and sport (on telly ;).

Tonight I'm planning on creating a template for documenting and sharing new ideas. I've been meaning to do it for ages, but right now I feel *just* motivated enough to make a start. If I really get going, I'll document up a new idea I've had around identity and "people search".

The Dip

Seth Godin's most recent book is called "The Dip" It tells of a dip in the adoption curve, and that it may in fact be a cul-de-sac. Good entrepreneurs known when a dip is actually a cul-de-sac and quit.

The dip I'm facing now is one of motivation. Real Job is sapping my energy at the moment. so I'm finding it very hard to push on through the last bit of MLTA development at night after a long day. The one thing I have learned is not to push myself though; things will get done quicker and better if I do it when I'm ready.

Facebook App update

The number of users of our Facebook app is growing. It's actually quite amazing when you consider it's had zero advertising. Of course, the numbers are still very modest (48 at time of writing), but it's encouraging. There is definitely a market there, the only question is how long it will last. Make hay while te sun shines as they say, so I think FB is the place to be for a little while.

Our app, "At The Moment ...", is basically some statements that need to be completed, e.g. I am drinking, I am working, I am dating, etc. It's not earth shattering, but it seems interesting. I have been looking through the database of entries and it's quite enlightening, a small slice of people's lives. It's also prompted me to do a little bit of work on the app rather than my usual Fri. night pastime of Xbox360 until after midnight :)

SCV site done

Well, no MLTA work got done on my 2 day break from Real Job, but I did manage to finish the website for Swift Current Ventures which is pretty cool in a day. I'm not sure if I've mentioned SCV before by name, but it's basically the "R&D" company and trust I started with my buddy in Qld that we'll use to run MLTA and other ideas.

My Facebook app, At The Moment ..., has 6 users (probably all my friends or bots) but it has already generated US$0.24 in ad revenue, which would be more than awesome if the user base grows significantly. Time will tell I guess. I also think that the revenue will probably not be sustainable considering it's coming from other Facebook apps advertising themselves. Of course, revenue from advertising was never the goal for the app, it's more about user numbers and a quick sell. Ad revenue is just an added bonus if it appears.

Right now I'm off to play a bit of Xbox360 (Blue Dragon ftw), and I should get back to MLTA on or before the weekend.

Finished (almost)

I've taken 2 days annual leave from Real Job to get some stuff done, and today I completed my Facebook app At The Moment ...

When I say complete, I mean it's perfectly functional and released. It even has som eheavy caching just in case it get's a major load. What it is lacking, however, is an easy way to share itself among Facebook owners, and there's also a stats page that needs updating. I think it will do for now, I really want to use my last day off to get the SCV website up and to finish off MLTA.

Still going

Some setbacks with my Facebook app means it's still not ready ... but it's very close. I was having drama's on the live web server. Most of the issues are because I'm using Java and Struts2. I could have taken the easy way out and run the whole app in an iFrame, but if in future I want an FBML app then I'd need to go through this pain, so I may as well do it now. Besides, using FBML allows me to leverage the Facebook look and feel, which makes the app look that much more professional and a part of the platform.

I'm thinking I have a day or two to go with this app, then I'll turn my attention back to MLTA to finish off a couple of bits and get it deployed for Beta. I also have to do a website for SCV and a few other bits and pieces associated with that. When those are out of the way, I might write an eBook on Facebook apps using Java, Struts2, Spring and Hibernate. Maybe.

Facebook-a-gogo!!

I've written my first Facebook app tonight, and after I worked out the kinks, it's all running pretty smoothly. I should be able to learn a lot from this, and so the next app I write will be a lot easier and quicker (though writing an app in one night isn't bad :). Of course, when I say app it's really a kind of simple portlet, with a component for the user, and then some text displayed on the profile page.

The Facebook api has been out a little while, though I've only recently got involved. I read on TechCrunch that a Facebook app recently sold for US$3m ... not bad!! I can see why though; if you write a useful app that gets a few million users, thats some valuable eyeballs to someone out there.

My app still needs a bit of polish, plus it needs to go on the live server, all of which I will do tomorrow night. I will probably also stick some Google ads down the bottom of the user page .. just in case ;)

Opposites

Starting up a tech business is my "night job". My real job is as a Solutions Architect for a software company, and the 2 roles couldn't be more different. Real job is slow moving, steady, and very large in scope, whereas being an entrepreneur is the complete opposite: you need to be very agile, small scale and fast to implement. The steady income is with RealJob, but there is way more upside with being an entrepreneur, even if a big payout never comes.

Starting up various ventures gives many personal rewards above and beyond anything RealJob could offer. There's the excitement factor, the feeling of being involved in a new frontier (Web 2.0) as opposed to "just another software system". There's also the immense feeling of satisfaction and gratification in seeing an idea make it all the way onto the web. Above all though, being an entrepreneur is about constantly challenging yourself, then rising up to meet it, rather than just plodding along doing the same things that were done 10 years ago.

What's in a name

Just a quick post ... we've been trying to come up with a name for our new umbrella company, and it's almost impossible!! The problem isn't cool names, it's uniqueness, which again goes back to the web domain name problem as well as a legal one I guess. Oh well, maybe my 2 year old daughter will utter something that will catch on ...

Company structure

We've decided on a company structure. A hybrid trust owning all the assets of MLTA, with us owning equal units in the trust. This way, we can sell off MLTA and not pay the usual company tax of 30%, but rather get taxed at our marginal rates. The reason for following this path is that the intent is to sell MLTA, not to earn an income from it. The most likely outcome (for any new web venture really) is that we generate a fair amount of traffic then get bought out by a large media company that can really leverage the advertising dollars. There's no point slogging away making a few bucks; the focus needs to be on quickly and cheaply building traffic and selling. The trust will also help us start any other businesses (I have loads of ideas :) and easily sell them too, without having to go through the hassle of setting up a company for each idea. This is especially cool because it reduces the cost of failure.

A subscriber!!

I've had my blog listed with Feedburner for a few weeks and have always come up with a big fat Zero for subscribers. However, Blogger has recently added some functionality to redirect every feed through Feedburner if the blogger so desires, which I have done. And guess what? I have a subscriber!! One, single subscriber. So. of course I am wondering which one of my friends you are ... so please leave a comment and say 'Hi' ;)

In other news, MLTA is pretty much ready for Beta. The idea is to target two very small areas in Brisbane, and see how things go. Once we've studied the numbers and feedback, we'll either be able to roll out on our own, or at least know how much we will need to raise in VC funds (or similar) to get penetration. Or, of course, we'll know if we've got a flop on our hands; but at least we'll know before we've invested tons of money we don't have.

Remember the ...

Ideas have been popping into my brain thick and fast these days, so much so that it's a struggle to keep going with MLTA. I am staying focussed however, devoting 100% of my at home work time.

I have started using Remember The Milk. It's a cool kind of ToDo list, all ajaxed up. I particularly like it because I can list my ideas down as tasks, and even add a little note, so that I can get back to them whenever. Two big plusses: All my ideas are in the one place (instead of onvarious bits of paper around the house and at RealJob) and I can access it from my mobile. This will let me jot in simple ideas, usually a domain name which reminds me of the idea. Brilliant, and of course it's all free.

I have also finished another Seth Godin book called "The Dip" which is as always excellent. It has really made me rethink not only my short term strategies for things, but also to take a slightly longer view. I'm debating the wisdom of staying in Australia; everything seems to be happening in the US. y wife and I would like to settle in Canada one day, so maybe a few years in Silicon Valley may be just what the doctor ordered. If MLTA is a flop, I'll probably push on to the US and try and get a RealJob over there before starting any new ideas, if it's a mild success we will probably use the money to move to the US and continue down the path on startups, and if it goes huge we'll move to Vancouver :)

Quick update

I've been ploughing on with the consumer facing part of MLTA. I was previously using the Dojo toolkit within Struts2 for AJAX stuff, but I've just ripped it all out and replaced it with my own custom code. The reason is this: the toolkit takes forever to load and initialise, and I really only wanted to use a couple of the features. My own code may not have 100% compatibility, but I guess that will come out in the testing. For now, I've probably dropped about 10 seconds off the load and initialise time. Not bad IMHO :)

Live ... ish

MLTA is up on the live server, so my business partner in Brisbane can test it. I'm still finalising the formatting and logo, but it should be all done by the end of the week. I've had a quick look at some performance tweaking too. There is a major piece of work to do to really save some processing time, but it's on hold for the moment unless its really noticeable in testing. Oh well, back to it!!

Functionally done!!

OMG!!! I've "functionally" completed MLTA. I poured in a lot of hard work (except for Sat. night when I decided instead to watch both the Wallabies and Socceroos lose) and now it's looking good. Well, it's looking crap, but it does everything so tomorrow night after Real Job I'm going to put the site up and let my business partner test it. Cool!! This is a real milestone, and I'm feeling pretty good. I know that there's still a lot of hard work ahead, but there's a solid base to move forward with now.

One big secret that I'd like to share is that of manageable scope. MLTA was in danger of never seeing the light of day for a number of reasons, not least of being the amount of work I had to accomplish in the evenings after Real Job. One big turning point was reading Seth Godin's nook (mentioned in another post). Another was when I decided to cut out a lot of functionality in the initial release. The best decision I made was to drop or delay a lot of bells and whistles and focus on the core functionality that will deliver value to MLTA's user. So, the most important lesson I've learned so far is not to bite off more than I can chew :)

When to release

I'm currently struggling over when to release MLTA. I don't want to rush it, but I also don't want to keep adding "just one more thing" and never release. I think what is behind this "crisis" is my desire to see some fruits of my labour, live and not just on my Mac. It would also be nice to start chatting about MLTA to other people ... the secrecy is killing me!!

What I am going to do is take a night off, and have a real think about what features I think are essential, and then go from there. Other planned features that aren't essential can always show up at a later date as a nice bonus for early adopters.

Structure

Just trying to work out what shape I want MLTA to take. I want to partner with a friend, but I also want to retain ownership of the IP (seeing as I'm creating it all). I have sent my accountant an email outlining what I think will work, so I am waiting on her feedback. I've read a couple of books while I was 'disconnected', and so now I have just enough accounting knowledge to be terribly dangerous to myself. I'm glad I have an accountant I can trust.

On another note, I've specced out an idea for another Web 2.0 site, and it shouldn't take long to knock up a prototype. However, I'm going to try and stay focussed on MLTA for now. It's nice to get the creative juices flowing on a new idea every once in a while during a project, just to stay fresh, but the challenge is not to get distracted and have 10 incomplete projects lying around. MLTA is my priority for now; the finish line is getting closer and closer, albeit slowly. Anyone want to kick in $100k so I can quit my job and work on MLTA full time? lol

Maths maths maths ...

I've just finished coding up some geographical calculations ... blech!! Now I know why I never go back and read my old High School mathematics books. Good thing I have Google, as it allowed me to search for formulas I needed very easily, and when I needed to solve a formula for a different parameter, Google helped me again and saved me ages of trying to remember how to do it for myself. I guess some people will bemoan the fact that it's easier to Google things than think for yourself, but hey, it saved me a lot of time and it's not something I want to be an "expert" in anyway. I'm an entrepreneur, not a mathematician :)

Ah ... connected again

Finally back and wired, and ready to start rolling MLTA. Logo work is proceeding. I think I'm going to be a real pain in the arse about it, but I want it to be perfect. The site redesign is going well, though it's still in what I like to call the "pre-pretty" stage. Once the logo design and colour is finalised, and the code is more or less complete, I'll begin the task of laying out and applying styles.

One thing that does have me perplexed is what charging model i should adopt. It's going to be free to use from a consumer point of view, because otherwise it would be a disaster, but from a "participating business" point of view I still have no idea how to structure. Should I have some free components and charge for "premium" services, or should I just charge a flat fee for inclusion. Decisions decisions, but I think it is going to be crucial to get right.

Where's the banana?

A nice side effect of not having net access is that I've been reading a ton of business books. One that has particularly stood out is Seth Godin's excellent The Big Red Fez. He describes a situation where (and I may remember this wrongly) a monkey in a big red fez is in an experiment. The monkey needs to navigate a series of rooms, and to get him going the researchers put a banana in each room. So, the first thing the monkey looks for in each room is the banana.

Now, Seth likens this situation to web surfers; the first thing they look for when they land on your page is the banana. Take Google for instance - the banana is right in the middle (it's the search box) which probably goes a long way towards explaining it's dominance of the search market. Seth uses example websites to point out all the things that distract users away from the banana and muddle up your site. Excellent reading, and invaluable advice IMHO. It certainly made me rethink my approach to MLTA, so much so that I have almost completely redesigned the page structure of the site so that you can't miss the banana. Thanks Seth :)

Still disconnected ...

But no longer homeless. We have found a nice townhouse and my office is setup. Just waiting on Internode to get my connection sorted out. They quote 7-10 days, which unfortunately means that I'll be in New Zealand when it gets connected. Hopefully when I get back, everything will be back to normal and MLTA and Idea #2 can get back into full swing.

I have got the ball rolling with graphics for MLTA, and have also done a fair whack of design work for the site. I'm just about to roll up my sleeves and start some code, but will need net access for some parts. I'm also a bit worried about my source code, as I am unable to use my server's version control through the bluetooth connection as it only allows port 80 it seems. On the flip side, buying a new phone and setting it up for the web will let me test "MLTA for internet-enabled phones", which was never something I had planned but will be an excellent addition.

Disconnected

Wow. We've been without internet access for over a week now and it sucks. I'm curently Bluetooth-ing my laptop and phone together so we can check email, etc., and I'm also sneakng in this post, but with a paltry 300Mb of bandwidth per month, we can't do much else :(

Development is on-again off-again as we're trying to find somewhere more permanent to live, as well as it being difficult to develop without net access. MLTA can move along a bit, but idea #2 really needs the net for research and reference, so it's stalled a bit. I've had some good ideas for it, but without net access to validate them it's pointless investing time. The graphics are almost done for it, but again are on hold until we can get full broadband access again.

All Packed

We're all packed up and on our way to Melbourne. This may be my last post for some time as I'm not sure about connectivity in the serviced apartment we're moving to.

Almost ready to launch idea #2. Some numbers: MySpace has over 100million users, Facebook has over 20million. Even if 1% of those users like my widgets, thats over 1million users. Multiply that by the number f hits each person gets and it's quite a large number of eyeballs. Hit on a great widget idea that everyone wants, and hold on for a wild ride. Hopefully I'll have connectivity and be able to launch the site in a week or two. At least I can do the development without the net.

MLTA hasn't had much progress, but a lot of what I am doing for idea #2 will be reusable by MLTA, so it's progressing 'via proxy'.

Progress

I've made some progress with idea #2 ... I'm no longer calling it my iPod idea because, like all good ideas, it has morphed slightly into something different but better. It will now focus on providing information widgets for webpages, with an iPod focused one first. I have completed the data model, some wiring, and some initial screens including login/out. Rego and then the widget are next and I'm almost done. I still need to contact the graphic designer to get a logo and colour scheme; it's much better to get things like that done professionally and spend a little cash than to try a home-made job and have the site look terrible.

I've also registered my domain name for idea #2. It's a made up word, but that's mainly because all the good domain names are taken. There definitely needs to be some change in the domain name area of the web, especially those that are 'parked' indefinitely. I understand the need for parking, but really, if you can't point your name at even a cheap hosted web page that at least describes what the site may provide, then how will you ever afford to complete a website? Of course it's all just in the hope of someone needing the name and paying the 'parker' a lot of money for it, but with the scarcity of good available domain names, something needs to be done IMHO. Rugby Union has a 'Use it or lose it' rule with regards to possession when in a maul, and I think the same should apply to domain names.

Mac attack!!

I've decided to move everything to my MacBook Pro. Everything. This will be the first time that I've ever done development on a Mac, and so far I'm loving it. Of course, the fact that I use Eclipse, Java and Tomcat (among others) to develop my ideas certainly makes it much easier.

I've completed the design work (on paper) for my iPod idea, and am now beginning the code. It should only take maybe 3 full days of effort, but with our move to Melbourne and sorting out the new job, that could be stretched out over a few weeks.

Moving?

Well, life has pretty much been on hold this past week+, as I'm trying to organise a new job in Melbourne. I flew down on Monday and loved it. The family is hyped and we're already packing, all we are waiting on now is a final offer.

So how does this impact my projects? Well, that's the beauty of the Internet I guess. There should be absolutely no impact, other than some time being diverted away to the move, etc. I'm loving the whole "virtual business" model, and I really don't see any reason why start-ups like mine need an office. Have laptop, will entrepreneur!!

On another note, I've come up with a good idea involving the ubiquitous iPod which should only take a few days to get fully up and running. I'm also aiming for MLTA to be code complete by the end of May. Time to roll up the sleeves :)

JSON enabling objects

So, MLTA is progressing nicely. In order to add to the "Wow" factor and maybe get some extra publicity once it launches, I've decided to fully AJAX enable it. The project is using Struts2, Spring and Hibernate (using JPA), and I've worked out a nice way of enabling AJAX calls via standard Struts2. What I've done is implement the JSONString interface, which basically defines a toJSONString method. You then build up the object using JSONObject(s). Any time you then want to retieve an object you simply call the standard find method and instead of returning a full HTML page, you simply use a JSP to output the value of toJSONString. Save's work out of the box, you just need to return a different vslue, for example 'success' for success, etc. Easy :)

Struts2 (formerly WebWork) includes some nice AJAX features out-of-the-box. One particularly nice feature is AJAX Tabs. However, one gotcha I came across was that you need all your Javascript outside of the Tabs. So, even for pages within a tab, the Javascript needs to be in the page hosting the tabs.

AdSense

I posted the following on Google's AdSense Group.

I think that AdSense is an awesome service but I think it currently lacks configurability from the content providers side. What I really think would be a useful feature, and someone please inform me if it already exists, is for content providers to be able to insert different "keywords" in the AdSense code to retrieve ads based upon those keywords, in a sense enabling "cross-sell" opportunities.

For example, let's say I have a website that sells widgets. I sell expensive widgets, and cheap widgets. As it stands now, I will get Google ads about widgets. What I would actually want is ads that appeal to different market segments, so I might want travel ads to appear on my expensive widgets pages and dating ads to appear on my cheap widgets pages. This could be accomplished by including the keywords "travel" in the code for AdSense on my expensive widgets, and "dating" for the code on my cheap widgets.

I'm sure that this change would not require much effort on Google's part, it could simply be an extension of what happens now with the search results on Google.com, but I think that this functionality will certainly increase the success of Google ads on many sites.

Here's hoping someone sees it, because it will really open up marketing possibilities on websites that I'm planning.

A meeting

I met with my very good friend Zej today, and he's agreed to come onboard MLTA. We're going to split it 50/50, with me originally doing the coding & technical work and Zej doing the testing & other stuff. Things are looking up for my first 'real' startup!!!

Google maps

Made a little more progress with MLTA today. Developed some cool AJAX stuff to lookup an address in Google maps. Very easy really, but powerful, especially for my idea. It will also make sure that there are 'proper' addresses in the database, not just rubbish. It uses Google maps Geocoder, GClientGeocoder, which parses a location from a string passed in. My code uses the returned JSON object to populate fields on a form which are then submitted to be committed to the database.

Development underway!!

I've started developing MLTA, using what I would consider "best practice" open source technolgies; MySQL with Spring, Hibernate and Struts2 for the front-end, running on Tomcat. I've also started using JPA - I totally love it!! Annotations are the best thing to happen to Java in a long while IMHO. I am not a big fan of writing XML configuration files, so I love it that I can use Hibernate JPA annotations on my model classes and it all gets wired up.

One super gotcha that took me a while to work out was that you can't mix where you put your annotations within a file. What I was doing was putting the @Id annotation above the member variable declaration, than a @OneToOne annotation above a getter. This caused a "Could not determine type for: xxx, for columns: [org.hibernate.mapping.Column(yyy)]" where xxx was the target of the One To One and yyy was the column affected. Once I moved the @Id annotation to above the getter as well, everything worked. I'm sure it's documented somewhere, but seeing as I'm teaching myself using web resources as fast as I can, I probably missed a rule or 2 somewhere.

I had a look at the Spring MVC component, but I think I'm going to stay with Struts2 (aka WebWork). Spring has some nice stuff, but I just can't go past Struts2's xml vslidation and rich taglib, which is miles ahead of Spring atm. I'll keep an eye on Spring MVC, as I'm sure it'll continually be improving, but for now it's Struts2 for me.

Plan complete

Completed my business plan for my latest idea (to be known as MLTA from now on). I sent it to a friend and we're discussing options this Sunday when he flies down from Brisbane. If he's interested I'll partner up and we'll try and do it on the cheap to start, maybe getting investment if it gets too big for us.

I'm working on the code all this week, trying to get most of it done. I've had a look at Google Maps API and it's a pretty cool app, and it's given my some great ideas (and functionality) to use in MLTA. Programmable Web has lots of Web 2.0 APIs listed, as well as mashups that use them. Great stuff.

Good lunch

I had lunch with David Bolliger of TILEFILE today. Very interesting. He's a pretty switched on guy, and he gave me some excellent advice. I've certainly got a lot to chew over.

I think the main thing I took away from the lunch was that I need to back myself 100%. I tend to look for validation from others a lot (even if I don't really care about it, which is weird). I also to focus less on being liked and more on getting shit done.

While lying in bed tonight, I've had *another* awesome idea for a .com, so I hopped right up and got the wheels rolling. I've registered the domain name already (you can never be too quick with these things I find), and will do the business plan tomorrow.

I've also made a minor modification to my Website Monitor program to add a 'Check now' menu item off the file menu.

Free software

I released version 1.0 of a web monitoring tool (Website Monitor) on my consulting site. It only took a few days to write, and I use it myself, but it's not something that I would ever sel. So I have released it under the GPL with the hope that someone will find it useful, tell others, and maybe drive some traffic/business my way. I'll definitely keep adding to the free offerings too, as I think it's not only a cool thing to do, it's also good business.

Search and web stats

I'm currently building a business plan template that I will use to try and get funding for my various ideas. It will be geared at pre-seed/seed investors with the plan running through to exit. I will probably offer it for free from my consulting site just to get some visitors.

Of course, in my research I dug up a few things that I didn't realise, or rather that I knew but never really pursued properly. One was that I need to submit my websites to search engines. I always knew that I needed to do it, but I never really dug around into the nitty gritty. The other thing I found out is that it is very difficult to get good web traffic statistics. Alexa's stats are skewed to those users who have the Alexa toolbar installed, and other stat sites use similar methods.

Finally, I've found it very difficult to get numbers for revenue from web advertising to put in my plan. The simple fact is that it's impossible to estimate how much traffic your site will get and what the click rate on your ads will be. More news later.

Mad month

Wow, what a mad month. With the full support of my better half, I quit real job ... it was just time. Scary, but so liberating. I'm feeling really happy and energised, I didn't realise how demoralising real job was. So now I'm trying my hand at freelance IT consulting (www.sjgoldie-consulting.com) and will also try and get some Web 2.0 sites off the ground.

On the Angel investor front I still haven't had much luck but neither have I been really looking. A friend of mine kows a company that is looking to invest though, so I will try and get a few proposals in front of them to see if they bite.

On the mentor front, I met an Aussie entrepreneur (at a job interview) and he's agreed to mentor me. Thanks David Bolliger from TileFile:) I have to set up a lunch next week sometime (hopefully). I just basically want to listen to his story and hope for some inspiration and insight. After that, hopefully David can help me with contacts and advice.

It hasn't all been up though. I've had to sell some shares in a company I helped setup, mainly because I got screwed by old real job. I had agreed with them verbally to do some contract work (about $10k worth), but after messing around for a few weeks, they backed out. What's really disappointing is, unintentionally or otherwise, they have basically got me to design it for free in the proposal. Live and learn I guess. Time will tell how they go, but unless I'm really strapped for cash I doubt I'll be working with them again. I have 2% of the company too, but no one to sell it to. Oh well.

New idea

I've decided to offer IT consulting services to start-ups. I'll initially keep it very small scale, advertising in a local Australian publication (Australian Anthill) with an associated website and see if I get any bites. My idea is to offer everything from advice to prototyping and web space using a very flexible fee structure. I'm thinking something along the lines of free initial consultation then either cash or equity for any major work. I'll have to squeeze it in to after hours as I can't afford to give up a real job (just yet anyway :), so I'm going to try and keep my involvement low but value high. We'll see how it goes.

Proposal

I created a short Keynote proposal for a friend of mine to see if he's interested in starting something with me. It's what I would call a "slow-burn" startup. There won't be a lot of initial effort, content will come in dribs and drabs, and I would expect site traffic to grow slowly. However, once it reaches a critical mass it may be worth a nice sum to the right buyer.

If he's not interested, I think I'll try and find someone else. My idea is to write the backend code myself, then have a partner code the JSP/HTML/Javascript. This will become much more efficient for me as I build up a small library of reusable compnents.

Back again

Holidays from Real Job have come and gone, and I have also taken the time to ease off all work related activities, as I badly needed the downtime. I'm back now, but only starting slowly, as Real Job does take a lot of my energy and I don't want to burn out.

Tonight I worked on an idea for marketing that can begin before a company is even formed, and that can be leveraged during negotiations with funders. Still not much to report on the Angel or Mentor front ... I'm beginning to think that it's maybe not going to happen in Sydney. Going to finish off the resume and send it out to a job I saw on seek. It's a step up from my current Rea; Job, and may offer some better networking opportunities.

Also discussed moving house with the wife, as we will no longer need a home office. Real Job is getting one of their own, so no more working from home, which is a good thing. Unfortunately, the location is in the suburbs, which while better for me to commute to (it's walking distance), networking opportunities will probably be nil.

Business Mentor

Finished my resume off, but I'm not so sure about the formatting. I have a long and short version, but the Pages template I used is a bit crap to be honest, so I still probably have an extra night of style changes ahead of me. Also watched Office Space with the wife tonight, which was a nice change of pace. I love watching movies with my wife, there's just something special about sharing a movie with the person you love.

Anyway, I'm currently researching business mentors. Angel investor's may be hard to find, but I think I should be able to find a mentor fairly easy. The challenge will be to find a good one. I think that before I go to a company that specialises in business mentoring, I will try asking a few people I know if they know anyone who might be suitable. I may even fire off a few cold-call emails to some successful Aussies, because you never know I might get lucky and pique their interest.

More resume work

Almost finished polishing my resume. I'm using Apple Pages, which is a very nice program, but I don't think I like the layout I've chosen. So I'll probably spend tomorrow night finishing up.

I've done some research on migrating to Canada too. Shouldn't be too hard, especially once I get some contracts under my belt. I spoke with a friend who has a resourcing company (plug: TRS), and he gave me some info on the ins and outs of contracting. Sounds pretty complicated, and I remember it was these complications that scared me off last time I considered it. Still, it's the best way to make some good money, and if I can move around enough, I just may stumble across someone who can help me get my ideas started.

Certification

Mainly worked on my 05-06 tax return tonight. Zzzzzzzzzzz.

I also had a look at updating my Java certification. Currently, I have Java 1.4 programmer certification. This is enough for me to do further certification, and I have been thinking about Web and Business objects certifications, which is basically JSP/Servlets and EJBs respectively. However, the exams cost $316 plus GST each, then there's the cost of exam prep books. I think I'll see how my contract hunting goes without these certifications ...

The Dreaded Resume

After another awful day at "Real Job", I've decided a change is in order. I think I will start looking for some comparatively low stress Java programming contracts to pay the bills while I try and get my ideas up and running. I'm finding that the stress of my current job, Technical Director of a very small software company, is really affecting my ability to function as a normal human being. Plus I can make more money contracting, which will give me some more flexibility. Also, my current job is a work-from-home situation, which was been great while ny daughter was a baby, but isn't exactly conducive to networking. I think it's time to get back into the real workforce.

Of course this brings with it a task I dread: updating my Resume. I actually have to do three ... one for the start-up focusing on all my skills, one for Java contracts focusing on my programming skills, and one for management contracts focusing on my management skills. So the next few nights I will be "massaging" my resumes. Fortunately, my current resume isn't too out of date, but I don't have anything on my current job.

Today, my wife picked up an application form for a post off box, so next week I'll get that sorted and then the business cards will follow. I have decided to go with Vista as they have a design I like and 100 cards only costs $12.49, although I am yet to see the shipping costs.

I still need to setup the Mac OS X coding environment, so I may do that tomorrow during family nap-time. My first sample Dashboard widget will be a slideshow for iPhoto, if it's possible (I have no idea at this stage). Otherwise, I may make a newsreader for this blog.

Quiet night

Haven't done much tonight. Found a site to get some business cards printed, but I want to get a Post Office Box setup so I can put that address on the cards instead of my home one. I have also downloaded the Xcode IDE for my Mac so I can have a crack at a simple Dashboard widget.

I noticed this story on Digg about Mozilla earning $50 million last year, mainly from proceeds of it's Google search and click throughs on paid Google ads in the results. The revenue model I'm hoping to adopt for all my start-ups will be based upon free content (mostly user-generated) with Google ads and referrals, so the Mozilla news kind of proves the model. Of course you will only make money if you can get very large numbers of users clicking the ads, which is where compelling content and user-empowerment (i.e. Web 2.0) come into it.

Some links

Tonight I've been reading Australian Anthill magazine to get inspiration and ideas. There was an interesting article on 'flipping' - basically just building a start-up to be sold. This is exactly what I have in mind, so at least I know that it is possible.

I think I'll also research Apple's MacOS X widgets. I've seen a few and I think it's a great vehicle for advertising web applications. I really like the digg newsfeeder - great for getting your content out there for the cost of a little development time.

Some links from my previous night's Googling:

• The Art of Raising Angel Capital
• The Executive Summary
• ATP Innovations
• Catapult
• VANTEC

Where are the Angels?

Last night I googled Angels in Vancouver, with some promising results (which I'll link here sooner or later). Tonight, I did the same for Australia, with very different results.

The top Google hits were "matching" sites ... post your idea and they will match you with an Angel. I really don't like this idea, and I'm sure it's not the right way forward for me. Seems too much like a dating service :) Other sites that came up were more VC type companies. I'm not after a board that I need to pitch to, I'm more looking for a business partner with knowledge and capital to get us off the ground. I think it may take some digging around to even get a whiff of an Angel in Sydney.

First up

So, I've started a blog. Something new for the new year.

Basically, this blog will be the record of my attempts to start up my own software/Internet company. I have ideas, but no idea what to do with them. I understand the general gist of starting a company, but the devil is in the detail as they say.

I work for a software startup company at the moment, but even though I am doing all the work, I only own 2% - which isn't really much (of course, I haven't taken any risks either). I live in Sydney, Australia with my wife of 5 years and our beautiful baby daughter, and I find it difficult to juggle work, family and then more work in the evenings (when mum and bub have gone to bed). The general plan is to get some Angel funding to cover my wage and company expenses, and to get up and running as fast as possible. I'm also going to be looking for an Angel who can mentor me, and who can help with networking. I'm not worried about the technology side of things, but I do know that there's more to a successful software company than just the software, and I also know my limitations.

We are hoping to move to Victoria in BC, Canada, so the first problem is whether to start up in Sydney, or to pursue funds to start in Canada. I'm thinking it will probably be in Sydney, as personal contact will be important in securing an Angel or two.

So, I have my ideas, I know what's needed to get the software done, now my first step is to research Angel networks in Sydney, and also to complete at least a skeleton business plan.

Of course, I have no idea if anyone who stumbles across this blog will even care, so I may just end up talking to myself ...