Some points on live video

I've been using a goodly number of live video services over the last month or two. They seem to have sprung up like topsy recently, evidently due to innovations made by Adobe with their Flash Media Server 3 and ancillary products, not to mention support by Akamai et al. Those very few of you who remember my Table vs Jetski startup idea will know that I have some interest in this area, and I have a few thoughts about the nascent industry.

I have used Stickam, Yahoo Live, LiveVideo.com, and even an Aussie site which was in painful alpha (someone remind me what the name of it was!) which was populated by lots of screaming Germans. I'm going to concentrate here on the services which allow multiple live video streams on screen at once, including the ability to join the conversation yourself. That knocks out places like Ustream and Seesmic, which are otherwise worthy of discussion but not in this post.

I'm not auditioning for Duncan Riley's job at TechCrunch so this will be somewhat half-arsed. Lets' go with point form.

- Flash has always remained a dark mystery to me, so I would love to know why some services decide to go with four of the smaller "viewer" windows, like LiveVideo and Yahoo Live, whereas Stickam goes with six. What's the upper limit, if any?

- Any developers out there in this space should be following the Watch KATG Live thread on the Keith & The Girl forums. There is some interesting stuff there about KATG's previous video streaming partner, PalTalk, and how paid-up subscribers of PalTalk have had to be dragged kicking and screaming into the Web-based client era, not just due to the financial investment they made in PalTalk accounts but also the technical limitations of the new delivery platform, i.e. Flash, and the move away from an intimate micro-community to a wide open free-for-all.

- It strikes me that the interfaces to all of these applications could do with a dose of modularity. What if a user wants a 3x3 wall of 9 mini-screens? What if a user just wants to view the main screen and maximise the text chat window? What if a publisher wants to show their own video but only accept audio from the crowd, in a call-in environment? Maybe I'm spoiled by NetVibes. These things should be possible, surely? They would be pretty difficult in Flash, I bet, but it will come.

- I suppose you have to recognise that Silverlight is coming round the mountain when she comes, but Flash is dominant in this area. Does it occur to anyone else that this is a tad weird? Why hasn't a startup taken this obvious opportunity to develop a competing solution? Have the days of new proprietary Web client apps gone forever? I remember the days when every week would see the announcement of some new "rich media" downloadable thingumabob for Netscape, complete with its own file format and brand new entry to a standards body. That doesn't seem to happen any more: Flash gets bundled with every browser and that's pretty much it. Good for Adobe, I guess.

- Is there an industry here? I mean, one that allows profits? It hasn't immediately been smothered by the tough love of the RSS/podcasting crowd, which means that it hasn't been infected with the Californian hippy bullshit about it being "all about the community, man". However, is it going to be possible to sell TV-style interstitial commercials into live streaming video? Would the audience wear it? I suspect it has a better chance of happening than with audio podcasting, if only for cultural reasons. It would take a startup with a lot of connections to make it happen, nevertheless. One wonders if anyone other than Google could pull off a video advertising business.

I am a firm believer in live events being an underexploited opportunity for Internet startups. It is what my only successful business so far has been based on. Live events, especially those that go for hours with constantly updating content, deliver a startup huge amounts of page views even if the audience isn't very broad. Live video should be the next huge thing on the Internet. I only hope some of the little guys can get on board before the GEMAYA giants gobble up all the gold.

Steeeeee-rike one

I coffeed with Ben Barren today in the comfy surrounds of Errol St - or perhaps it is more accurate to say that I became immersed in the Ben Barren Experience. More than anyone else in my travels meeting the best and brightest of the Australian Web 2.0 scene over the past four years, Ben embodies the entrepreneur in my mind: gregarious, inquisitive, disarming, irrepressible... a creative mind so rich with imagination that he can't stop talking for fear of the spigot in his brain drying up. It is exhilarating to talk shop with him.

It struck me, as we swapped war stories of pitched PowerPoint battles across parched McCubbinite business landscapes, that it is a tragedy that the current Australian industry is not structured in such a way that people like Ben and myself, and others like us, can find an easy way to work together towards a common goal. Like the spinifex tussocks that my dad and I (mostly Dad!) used to have to attack year after year on our 20 acres outside Seymour in my childhood, Australian Web startup founders have to grow resilient and spiky to avoid getting consumed by the inexorable hunger of our oligopoly-dominated economy.

Many of us strive individually, feebly watered in various tenuous hierarchies by money men, activated actuaries and impatient investors. While it is true that startup founders need money, founders also need a creative environment in which to flourish. Something Keith Malley of Keith & The Girl said the other day about relationships is pertinent, even in a business sense: when you start dreading the knock on the door by your partner [personal/business/whatever], then you should know that it's not working and you should get out.

Thus we come to the subject of this post: my co-founder Tai Tran and I have decided to dissolve our partnership. This means no more Tinfinger, and it also means that I'll be the sole operator of FanFooty in future (handshake deal, a la Joe Wilson and the Galletly brothers). Yes, Tinfinger has joined the deadpool, become epic fail, et cetera: write its obituary up for your blogs if you care to. I feel confident enough to confront that fact head on, unlike some recent ex-founders, because I can hold my head up high and say that I have already had one success with FanFooty, so a failure doesn't mean I'm worthless. As I've said before, I'm following the Hitwise template of the cash cow followed by the home run play. I guess I didn't hit a home run this time, but I hear they give you three strikes before you're out. ;)

I should also say that I will be proud to have Tinfinger on my CV. It is a fine application in a technical sense: it works, there aren't any major bugs, and it had a rather high degree of difficulty to build. I learnt a lot in building it, and it contains a lot of elements that may prove useful in future efforts, like proprietary spidering scripts. It could even prove useful if someone bought the code off us (email me for inquiries/offers at m0nty dot au at gmail dot com) and put the elbow grease into it that we can't.

I have to admit, though, that I got a number of assumptions on the business/marketing side of the project wrong. The first of these was that the virality of Wikipedia would just "happen" for us too. I'm sure both Jimmy Wales and Larry Sanger would tell you that it requires a lot of work from a lot of people to become that viral, and we're just two guys in a house in Geelong. (I'll try not to use that excuse too much.) Our second mistake was that we entered a market in which, apart from the unstoppable juggernaut that is Wikipedia, we were competing with Squidoo, Spock and Mahalo, all of whom have major venture capital backing and roomfuls of employees. The industry, such as it was, evolved over the length of Tinfinger's development time into a fairly blatant Google SERPS spamming subculture, Frankensteined by more cash than we could shoot our popgun at. In particular, this Valleywag comment by Mahalo founder Jason Calacanis rang warning bells in my mind. Jason's back-of-napkin calculations meant that to keep up with his pace of bootstrapping, Tinfinger or any other small startup in this new ego-arbitrage space would have to be even more unscrupulous about middlemanning the barricades of constantly updated content that we didn't own.

And we were just two blokes in Geelong. (Sorry.)

So it's back onto FanFooty, and whatever else the future may hold. I don't really know what is next. I have a few crazy ideas at the back of my head, but they all require knowledge I don't have and resources I can't tap. I'll have to be satisfied with milking the cash cow for now, and looking for the next opportunity to step up to the plate and swing for the fences again. Hopefully next time I'll get a bigger piece of the ball.

Just to say I blogged this month

Let's keep this short and snappy. I haven't got the personal bandwidth to compose the sort of 1000-word essay that my usual blogging style requires.

I don't visit Techmeme any more, haven't for months and months. I get my news from NetVibes and Twitter. Haven't got the energy to keep up with the melodrama. Important memes have a way of percolating through.

One of those is the mystic rediscovery that AdSense isn't the Alpha and the Omega (linked because of comment #5). Well, maybe it's the Alpha. The Omega, apparently, is becoming your own advertising network. We've at least got past the Alpha stage with FanFooty, having ditched AdSense in favour of a local agency which uses DoubleClick. Still part of the GOOG family, I guess. What would you call this stage... Mu? The cow analogy works, the eatin' is good.

It's interesting that we were headhunted by this agency to replace another company in our space which just graduated to Omega status, leaving a big hole in their inventory. We're still just two blokes in a garage, operationally, while the Omega-bound competitor rents a North Melbourne hothouse with more than a dozen employees.

Tai doesn't like me lifting the skirt like this, but I think no harm done. Now, how long before I blog next time?

Twitter and IRC: meet the twain

As Richard Giles has already blogged, the Australian Twitterati have been gathering for the past little bit at the #twitter channel on the Freenode IRC network. This caused some interesting discussion on today's edition of the 2Web Crew, starring Richard himself.

Like Richard, I can't see much conceptual difference between IRC and Twitter. Twitter has a 140 character limit, IRC has about a 410 character limit before it inserts linebreaks. Twitter has a gateway to SMS and IM, but there is no insurmountable technical hurdle for anyone who wanted to do either of those for IRC. Similarly, it would be trivial to write an IRC bot to convert links into tinyurl URLs. Twitter has canonical identity, but IRC bots can enforce unique nicks with password-protected logins.

Then you come to the advantages IRC has over Twitter. Twitter is only now catching up on the idea of channels, and it's even stealing the # prefix from IRC. Most importantly, IRC has already encountered and solved the scaling problem many times over, with multiple servers all feeding the same chat space and netsplits handling server failovers.

You could pretty much replicate Twitter in a distributed format by setting up a tightly-securified global network of IRC servers, writing some AJAX/Java/Flash/Silverlight/whatever to pump it through, and then putting a Web front end on it all. If architected correctly, the Web site wouldn't fall over nearly so much as Twitter because the bottleneck would be on the IRC end, which would naturally have many backups. Server outages would be handled as IRC netsplits, which would (hopefully) be invisible to the user. The whole thing could be handled without relational databases given some tricky work with flat text log files.

Does anyone have the smarts these days to do that and make it work? Or would they rather roll their own db and architecture and not have to rely on a clunky 90s-era technology? IRC is like a library of reliable code, IMO. It would be foolish to not at least consider retrofitting it for a new application which is, at its heart, just a redeployment of the original IRC concept with a few more bells and whistles. I wonder if the Dave Winers of this world, who have been agitating for a distributed Twitter alternative, are capable of delivering (or funding) this sort of system.

Time for Aussie 2.0ers to Hitwise up

Ben Barren broke a blogging silence of three months today with an uncharacteristically lucid mini-essay detailing his thoughts going into the fourth year of Feedcorp, the business he started with consiglieri Michael Corleone but which is now headed by Pete Burley. Ben pledges to concentrate more on Gnoos, something which I have been urging him to do for some time now (perhaps too pushily?). Judging from the numbers he mentions obliquely in his post he seems to be doing okay with Feedcorp's hired goons strategy.

The Australian corner of the industry is doing just fine. We're in our fourth year (first lines of code on 27 October 04) as is Feedcorp, and Cameron Reilly noted that The Podcast Network officially turns 3 today too. TPN is going gangbusters according to all reports. Norg Media is expanding to new cities, Scouta TV is doing deals, Tangler is growing with a new CEO as well, etc etc. We're already running at over 100% year-on-year growth on last year's stellar traffic figures for FanFooty, and with yesterday's launch of a $5,000 fantasy competition (using all our own money) called Lethal League we should have an awesome 2008 where the growth of Tinfinger will be gravy, not our staple diet.

Notwithstanding the recent bad news about Omnidrive, there doesn't seem to be a lot of negativity in the Australian Web 2.0 community. It's all systems go from my perspective. So is the gentle backhander that Ben delivers to his peers warranted?

Knowing how hard it is to do (build a valuable company that creates more dividends individually than a job/freelancing etc or is acquired etc + being able to scale/keep your site up/pay bills/keep investors happy etc) I only feel empathy for the other Aussie startups trying to do it locally, but I wouldn't say I'm too optimistic on the likely 'home runs' with far too many consumer only/google ads/low cost plays, that are still pet projects not scaleable businesses.

This is a longstanding, if cheerful, debate that I've had with Ben: me on the side of sacrificing short-term profitability to spend long hours of drudgery building your own IP, him on the side of getting some B2B moolah while the getting is good from dumbshit oligarchs who need Remora 2.0s to sandblast the barnacles off their rotting underbellies. At this stage I don't doubt his numbers would most likely trump anyone else's in pure turnover terms (save for Atlassian, but they shouldn't really count in this context as they're on another plane already).

When it comes down to it, I don't see any approach as being "wrong" unless you fail, as it appears (from admittedly second-hand hearsay) that Nik Cubrilovic has done, sadly. From the sounds of it Nik just ran out of money, which is a win in my book for the "low cost play". Some business concepts have a greater ceiling than others, but then again the higher the ceiling the greater the potential for a GEMAYA player to lay the smack down, as Google recently did to Topix (and maybe Gnoos as well?) by adding local search to Google News.

I remember sitting down with Ben after a speech he gave last year and discussing the Hitwise model, which is to build a limited-growth business first and then use the solid cashflow to fund the slow-burning development of the "home run" concept. That is what Ben has done with Feedcorp, where his outsourcing work is the Sinewave equivalent and Gnoos is his Hitwise. In my case, FanFooty is the cash cow and Tinfinger is the home run, although FanFooty is obviously on a different level to Feedcorp in terms of revenue. I think this is the most likely strategy for successful Australian Web 2.0 ventures, because the first business not only gives you funding, but also the experience and confidence which are invaluable in making the second, harder concept work.

I look forward to seeing what Ben does with Gnoos in 08, and after launching Lethal League yesterday I will now be getting back to Tinfinger. I've given myself and Tai until October to make a living out of it. Next one to run out of money is a rotten egg! ;)

All your browsers are belong to X-UA-Compatible

An article this week on the Web design bible A List Apart lets us know the latest plans by Microsoft to embrace and extend the way HTML is rendered in Web browsers. Apparently in consultation with ALA boffins, Microsoft has agreed to implement a new meta declaration in the head section of HTML documents in their forthcoming Internet Explorer 8 release.

Using a simple meta declaration, we can specify the rendering engine we would like IE8 to use. For example, inserting this:

<meta http-equiv="X-UA-Compatible" content="IE=8" />

into the head of a document would make IE8 render the page using the new standards mode. This syntax could be easily expanded to incorporate other browsers as well:

<meta http-equiv="X-UA-Compatible" content="IE=8;FF=3;OtherUA=4" />

This is a great idea for developers, because they get to write once and then never have to edit their code again, no matter what new browsers are released. Never again will a new version of IE or Safari break their lovely site. Unfortunately, it sucks for users, mainly because participating browsers will have to bloat out to humungous sizes because they will have to include the rendering engines of all previous browser versions in order to be compatible with this new system. It also sucks for Mozilla, because part of their marketing message is that Firefox is the cleanest, smallest browser out on the market, and the inevitable bloat will blow that claim out if Firefox implements this system.

Microsoft caused this problem in the first place by not adhering to Web standards in previous versions of IE. Now they are trying to apply another band-aid over the suppurating wound, and they have enlisted a surprisingly self-serving ally in the ALA crowd. I would have thought ALA would be better than that. Developers should focus on developing standards-based code, minimising their use of browser hacks, and lobbying Microsoft to adhere to standards, not to cover their arses.

Flickr CC-BY attribution

I have fixed a problem on Tinfinger relating to photos uploaded from Flickr which are marked as Creative Commons licensed requiring attribution. As per the Flickr community guidelines, Tinfinger now links to the photo page instead of the source URL. Thanks Stuart Hamilton for putting me on the straight and narrow!