Fishpool

MySQL 5.1 continues community release madness

Osma — Tue, 02 Dec 2008 11:35:00 +0200

So MySQL 5.1 is declared GA, contains enough bugs to make Monty rail against it publically, doesn't (as far as I've been able to gather) have a stable cross-release pluggable storage engine ABI (despite that being a major release feature -- meaning you have to have new plugin releases for each MySQL server release before upgrading anything), and continues the madness of less frequent community releases and less stable enterprise releases against the wishes of every sensible enterprise customer and (I would guess) every community member.

I know enough about running MySQL to know that Monty's comments don't necessarily reflect much on the practicalities of someone trying to run something using MySQL rather than developing it, but the rest of the noise about this release definitely makes me want to seriously reconsider a policy of using MySQL Enterprise version and look at a real community-released version instead for production use.. There's just a "small" question of figuring out which community..

Mini Friday is not Habbo Hotel, but we're proud of both

Osma — Wed, 26 Nov 2008 17:43:00 +0200

This had to happen, didn't it? VentureBeat picked a slightly unfortunate headline for an info tidbit shared by Sampo and Aapo at Slush Helsinki about the status of Mini Friday, our mobile research platform. Now it's getting picked up by various other blogs, and even ended up at Washington Post with a headline which is not at all hard to read as if it talks about Habbo, not Mini Friday.

For the record: Habbo has over 114M registered accounts, while Mini Friday has 1M. The latter is a huge achievement on its own if simply for the fact we've never advertised it at all -- the whole point of the project is to see who would find it own their own, and what would they do with a mobile virtual world if there was no commercial pressure driving it in any particular direction. It's done that wonderfully, by the way. Still, Habbo is on a scale of its own, and pulling ahead at 3 million registrations a month. We have no intentions of trying to reach that with Mini Friday. An entirely different experience is in order for mobile use..

PS. It's not my intention to keep up this pace of blogging, so if you're subscribing and getting tired of these entries, despair not - they'll get more infrequent soon :)

Helsinki Airport shows a clue about wireless networks

Osma — Tue, 25 Nov 2008 21:01:00 +0200

..by launching a network which requires no pre-authentication. In addition it's free, which is also cool, but far cooler is that unattended devices (like syncing email without launching a web browser) work thanks to requiring no auth. That should go some ways to help those stranded on the airport during snow storms like this Sunday.

New reward models for Habbo

Osma — Tue, 25 Nov 2008 10:38:00 +0200

As I hinted at last Thursday, and as was noted by Sulka yesterday, we are introducing a second currency to Habbo. We've had Coins (or credits) since the beginning of the service as the purchasable in-game currency, and the business model of Sulake is primarily based on sales of this currency to end users via a variety of mechanisms and sales channels from premium SMS billing (our original method) to credit cards, PayPal transactions and prepaid voucher cards purchasable from kiosks and stores like R-Kioski and 7-Eleven around the world.

Since yesterday, starting from our pilot site Habbo UK onwards, there's now a second currency as well, called Habbo Pixels. This one can't be bought -- you have to play Habbo to get it. There's a whole new set of cool things you can do with this currency, that can't be done using Coins. Naturally, the process doesn't end here, and we'll be introducing plenty more features tied to Pixels, Coins or both over the following months as part of our routine monthly releases.

Because of this tight connection to our primary business model, and because trading is such a big feature of the gameplay in Habbo, the implications of this change are pretty substantial. And so was the process by which we arrived at making this change - starting from (at least on my part) utter confusion of why anyone would want to complicate their economic model by introducing this big new variables. Once we got that part (thanks to everyone who patiently explained why it makes sense), implementing this still took its own sweet time, as did all of the pre-analysis on why it wouldn't immediately collapse our end-user sales and drive the company out of business -- that it would be fun for users was less of a worry.

I've been for a while convinced this will be huge for Habbo. It's way too early to tell whether that will truly be so, but the first indications sure look promising. Our UK service had a new peak simultaneous users record on the day of the launch of Pixels (20% increase -- if you care, you can follow those figures on the front page of each Habbo site), and the community feedback is overwhelmingly positive, despite its normal bias towards resisting change. We'll be following this closely, and follow-up articles are sure to appear in many places.

High-profile shutdowns and low-profile major updates...

Osma — Thu, 20 Nov 2008 22:24:00 +0200

Quite a day for negative articles on Google. I counted 10 articles on Lively shutting down, with the usual suspects gloating and cheering from the sidelines. Won't be joining that crowd, it's never fun to see someone's work being thrown out the window. I did want to relate that to a bigger picture, though.

This certainly won't be the last shutdown among the 100+ virtual worlds projects under work right now, and I wouldn't be surprised at all to see even higher-profile projects being canceled, either in the current financial bloodbath which curely will kill even some companies which 6 months earlier were entirely viable operations, or in the year to come when the business fails to materialize. Money's still being thrown in to the game at pretty amazing rate, and there's just not enough experience to go around to make all these projects work.

And that's because at the end of the day, this business is not about the brand, the IP, the coolest technology, or even the best user experience. Instead, this is about being able to nurture a community, co-operate with it to develop something which no one knows where it will ultimately end up at, and to be on the pulse of what's going on, every minute of every day. I know it's easy to forget that, with the allure of focusing on the superficial, easy to analyze product features, APIs, and so on.

We've certainly learned that the hard way ourselves - it's not like we've always kept our eye on the ball either. Still, you only need to be right most of the time, listen carefully, and not miss where you need to correct yourself. It's now been roughly a year since we launched a significant refresh to Habbo that put the service on the track its been on since, after what can be described only as a pretty horrible summer 2007 for us.

That was 12 months and 11 upgrades ago, though. Yesterday we launched what's internally called release 28 of Habbo in UK. The most exciting features of it haven't yet been turned on, as we're preparing yet another global roll-out to begin next week. This is an important release to us, possibly as important as the one made a year ago, as it'll change the economic and reward models of the community forever. I hope we got it right on this one. If we did, it's going to be an amazing Xmas, for us and for all the Habbos. If not, well, then we'll need to scramble a bit to make it fun anyhow. But that's what we're good at.

Looking for a ETL engineer for our BI team

Osma — Wed, 19 Nov 2008 10:57:00 +0200

So, I mentioned earlier that I was looking at Infobright's Brighthouse technology as a storage backend for heaps and heaps of traffic and user data from Habbo. Turns out it works fine (now that it's in V3 and supports more of the SQL semantics), and we took it into use. Been pretty happy with that, and I expect to talk more about the challenge and our solution at the next MySQL Conference in April 2009.

However, our DWH team needs extra help. If you're interested in solving business analytics problems by processing lots of data and the idea of working in a company that leads the virtual worlds industry excites you, let us know by sending us an application. Thanks for reading!

Chris Anderson on freemium conversion

Osma — Sun, 16 Nov 2008 13:38:00 +0200

Chris Anderson, author of The Long Tail, uses free-to-play web games as a case study on conversion rates for freemium products. I wrote about the conversion and monetization rates in this world two months ago as a followup to my GCDC presentation from last summer. I can't really think of a better example of freemium model than Habbo - a freely accessible service with high engagement and a large audience really gets to utilize and showcase the model at its very peak. The only thing missing is even easier micropayment models. We'd love to use the iTunes store for selling Habbo items, for example.

Is "Marketing" and "Product design" the same thing?

Osma — Wed, 12 Nov 2008 20:57:00 +0200

Reading Raph Koster's thoughts on marketing in games vs the web, I can't help but be slightly confused. I don't disagree at all with the basic statement - that marketing (on the web the same as anywhere) is primarily about understanding what the customer wants, and that product designers should also be all about that. Nor could I ever disagree that metrics wouldn't play a huge part in providing that understanding - after all, that's what I've been preaching myself for as long as I remember (though only a couple of years in terms of customers).

I guess what I'm confused about is: who could have ever thought it could be different? Has anyone ever met a product designer or a marketing specialist who didn't at least claim to understand who the customers are and what they want? Now, you can of course say some of them don't know what they're talking about, but I could say the same thing about many members of other professions as well. Of course their job is to understand the customer's desires and come up with the best ways to address those in a way that generates healthy business.

And that's also where the professions diverge: yes, there's plenty of overlap as well, especially on the web where every action is nearly instantly measurable, and many product changes are trivially implemented. Even so, talking to both marketing and product design people (and not being an expert in either myself), I see a whole lot of difference in the approach and tools for the activities as well. You could try to do both at once, but apart from very small projects or businesses, I can't see how you could possibly spend enough time on both subjects to really produce the best results.

No, they're not the same. Not on the web either.

Splitting the virtual worlds market to segments

Osma — Thu, 16 Oct 2008 11:01:00 +0300

IMVU founder Eric Ries commented on Virtual Goods Summit and suggested that virtual worlds can be divvied up along three axes of UGC/first-party, subscription/pay-for-stuff, and economy/gameplay focus. This is certainly one good way of thinking about the focus decisions needed when designing and developing a product in this market, but personally, I think this model, along with others I've seen and played with myself, suffers from a few key weaknesses that arise from the need to simplify things. I'm not saying the model can't help put things in order, just that there's more to finding the right solutions than this. Lets go with the great blogger tradition of point-for-point response.

UGC vs first-party

It's amazing Sulka didn't comment on this: UGC is not just about letting users upload pictures or items to a world. More to the point, Habbo certainly is not first-party content focused. Yes, all our furni is designed and developed by our own teams, and we don't enable user uploads. But at the same time, over 90% of all of the activities in Habbo emerge from the community - users take what we've made, and do their own things with it. Most of what's going on, we had no idea would happen.

Eric says IMVU's efforts to enable UGC dwarf those to create their own first-party catalog. Well, so do ours, despite his classification of Habbo being first-party content focused. Every feature, every furni, every activity, every news item receives more thought on "how do we support users to go to their own directions here?" than "what do we want this to be about?". Plus the significant fraction of our work that has absolute no effort to produce content attached to it, and is fully focused on player activities.

Lets just use the old, tired LEGO analogy here. How much of LEGO is first-party content? Just enough to get the imagination of the players going so they can create something of their own. Anything more would be too much, and this applies to any VW that can call itself "social" - and none that isn't social isn't going to be interesting. Trying to make a useful UGC split for any purpose other than copyright infringement monitoring is a red herring, and even for that one purpose it's not very likely to be useful due to other moderation requirements.

Subscription or pay-for-stuff

This is one of the stronger arguments, if only just because those are the business models the industry has latched on to. They're certainly not the only possibilities though, nor are they alternatives to each other. Eric's points about the strenghts and weaknesses are good - but you can benefit from both at the same time, and support the weaknesses of one model with the strenghts of the other. This is certainly an area where we have a lot of experience, over 8 years of it, and I don't think we've gotten very far yet..

Economy or gameplay

Eric used the word "merchandising" instead of economy, and I think that's the crucial over-simplification that leads to thinking that pay-for-stuff games and worlds are just about cross-selling opportunities best left to a competent marketing department to handle. I'm wondering whether he's simplifying the choice to make it easier to explain, or purposefully misleading someone on what's crucial to think about, or whether our friends at IMVU simply haven't realized this yet: the first-hand sales are a small fraction of the total trade in an item-based game, and the gameplay balance is just as critical here as it is in a game built out of designer-created quests and gameplay mechanics. What's more, because its emergent behaviour, it's nearly impossible to predict, and very difficult to measure, model and understand. Yet that's exactly what's required in order to succeed.

I hope that explains why I choose to call it economy-driven rather than merchandising.

PS. Browsing around Eric's blog a bit further, this article is a gem

Infobright BI tools go open source

Osma — Mon, 15 Sep 2008 20:27:00 +0300

I've mentioned Infobright before as an interesting solution to getting more performance to BI analytics solutions. Today's news are interesting: Sun invests in the company, and the baseline product is open sourced. Too busy to write more about it today, but I'm certainly watching this one closely.

The sweet spot in free-to-play, pay-for-stuff market

Osma — Fri, 05 Sep 2008 13:53:00 +0300

I've been talking recently about a few particularities in the business models based on end-user micropayments that have created lots of followup discussion and questions. So much, in fact, that I decided it's time to try to explain one crucial and somewhat counter-intuitive detail in writing for later reference.

First, a bit of background: this information is based on my work with Habbo over the last 5 years, and is half learned from experience, half based on theoretical models built from that experience. I'm sharing this with the world because while it's been an interesting ride to build an online social game with an end-user business model, breaking pretty much every conventional rule in the process ("games have to have objectives", "there is no profit in micropayments", and so on), it's still better for our business if people understand why it works. If this allows a competitor to fix a problem in their product and get off the ground, so be it - there's plenty of growth to go around here, and failures don't help anyone. As a disclaimer, the numbers I'm discussing here have no relation to Habbo, though the basic observations certainly apply.

Let's start with an obvious statement and follow it up with something less obvious: Everyone wants to maximize revenue per player. However, in a free-to-play environment, where the majority of players do not contribute direct revenue, the right tool for the job is not to try to extract the maximum amount of money from those who do pay - rather, to increase the number of players buying anything at all - even if it's just $1 over their entire lifetime. In other words, it's good to have a lot of very low individual value players.

To explain it in detail, lets look at two assumptions behind a flexible pricing business model: first, that the number of customers grows as the cost of goods drops, and second, that the maximum consumption is unrelated to the minimum. There is no average customer who would spend more than half of others, and less than half of the rest. If there were, the picture of that customer base would look something like the image here, and it's pretty strange looking, wouldn't you say? You've probably seen pictures resembling this one where they don't start from the dominating $0 value point - that's the normal distribution.

The first assumption really is very simple: more people are willing to buy a product at a lower price. This is true for most goods, with some notable exceptions in the luxury goods market, where the perception and desirability of a product goes up with its price. However, it is difficult to create a mass-market luxury item, and those do tend to be cheap (and small).

The second is perhaps slightly more involved especially if one is used to thinking of fixed-price models such as one-time purchase of a boxed product or monthly subscriptions, both of which are difficult to scale up on a revenue per customer basis, so scaling them down is highly undesirable as well. However, it's more clear, if not obvious, by looking at other consumer goods - whether tangible such as drink- and foodstuff or intangible like movies, music and other entertainment. Buying these once certainly does not exclude further sales of the same product to the same customer - rather, it's a strong indicator of sales potential!

The free-to-play, pay-for-stuff model follows both of these assumptions. Cheap purchase price attracts more customers out of the existing free users, and transactional item-based sales allows repeat purchases of theoretically unlimited amount. Those who are willing to buy more will do so, up to some practical maximum of consumable goods and discretionary spending.

In this environment, focusing on higher-paying customers makes sense only if the number of customers drops by less than half when the revenue per customer doubles. Again, with the exception of some luxury goods segments, this rarely happens. Think about it: how many chocolate bars of standard quality would you expect to sell for $1? How about for $2? More or less than half? How about for $10 for the exact same package? I'd wager chocolate bars sell at least 10x better at the price of $1 than at the price of $10 each, and the increase of customer base more than covers the lower per-unit revenue.

This is a simple exhibit of power-law market dynamics, and is easiest observed when looked at through a logarithmic chart. Readers of books like The Long Tail or Critical Mass should not be surprised. There's a twist through - because this starts from zero gains (at the free players), the exponential behaviour follows a different path in the beginning. This model also turns Pareto's Law on its head - due to the (in my experience) relatively high exponent, the highest total value is at the lowest end of the spending.

Now, of course there is a minimum profitable price for a bar of chocolate that does not become near-$0 even at very high volumes, unlike purely digital products, so increasing chocolate-sales revenue by dropping prices does not necessarily increase profits, and I'm completely ignoring the effects of packaging and marketing on the perceived value of items. For digital sales, where packaging is more flexible and material costs are effectively non-existent, we still have to consider not-unsubstantial fixed development costs, a certain amount of costs associated to servers and bandwidth, some transaction-related pricing friction, and so forth, but certainly the minimum value (and price) of one unit of digital sales can be driven much lower than a bar of chocolate.

Leipzig GCDC next week

Osma — Mon, 11 Aug 2008 23:07:00 +0300

So, I've been working all evening on my presentation for next week's Leipzig Games Convention. Its title is the same as the one I had in ION last spring, but I keep digging up this new interesting stuff that just forces me to rewrite everything. I'd never make a great consultant because I wouldn't be happy reusing what I did for the last client. I guess that's why I love the product world instead :)

Anyhow - I'm bummed because I was planning to stay in Leipzig for the entire conference and first day of the expo to meet people and learn new stuff. Instead, it turns out I need to be back in Helsinki on Tuesday, so it's going to be an in-talk-out type of experience - precisely the kind of thing that does not make any sense for conferencing. Can't help it this time, though. If you're at GCDC on Monday though - let me know, would love to make the best of the time available and maximise the people to meet anyway.

Virtual worlds looked lively today

Osma — Wed, 09 Jul 2008 21:55:00 +0300

This turned out to be quite a day for virtual worlds' publicity, with both Google's Lively and Vivaty Scenes launching (congrats to both), and Second Life annoucing an interop with OpenSim together with IBM. All of that is great, and validates the concept of web-based, interoperable virtual worlds supporting and interacting with other web content (and vice versa), and supports our own thoughts nicely. There's plenty of room here for more players! I'd love to write more on this, but with my last work week before summer vacations keeping me quite busy right now, I'm afraid I don't have the energy to put my thoughts down to a publishable shape. Perhaps later, though. I'm sure I will have a lot of interesting discussion about the whole scene in GCDC next month, as well. Look me up if you're reading this and plan to be there!

(oh yeah, sorry about the horrible pun in the title)

100 million and other metrics

Osma — Fri, 27 Jun 2008 17:32:00 +0300

News hit the Internet this week that our favorite social play phenomenon, Habbo has reached a milestone of 100 million registered characters. Several places in the blogosphere have also pointed out the weaknesses of that figure, such as the fact that yes, most of those characters are "alts", abandoned accounts, or otherwise not very meaningful. I'd be the last person to argue against that particular point.

Still, it's a figure that has some comparison basis. MySpace didn't have 100 million registrations when News Corp acquired it, reaching that milestone a year later, and MySpace regs are arguably just as likely to be abandoned as virtual world avatars.

Overall though, we don't really put much weight to that metric compared to many others. The 20 million new registrations made in the last six months is pretty nice, but nicer still is 1.5 million logins per day and over a million hours per day spent in Habbo by users. Those are hard to discredit, never mind whether you think that free-to-play is less sticky than subscribed games or not. I tend to give more attention to that than to the 9.4 million unique monthly visitors, because ubiquitous as the UB/UV measurement style is, it, too is horribly inaccurate due to all kinds of errors from cookie washing to shared computers.

If you're interested in hearing about how we measure Habbo and decide what to do based on what we learn from those measurements, you might want to consider attending Leipzig GCDC this August, where I'll give an updated and more in-depth version of the talk the slides of which are in my earlier posting.

Social gaming basics

Osma — Fri, 06 Jun 2008 10:58:00 +0300

There's a good article over at TechCrunch about the newly discovered phenomenon of social gaming. I mean newly discovered in the sense of widespread game industry recognition, of course - Habbo was reaching more teens than any other teen media (TV channels and popular magazines included) in several markets already years ago, while most so-called industry experts claimed the model can not work. We didn't come up with the terms social play and social gaming, though - unfortunately we called in free-form play in a social context, or something equally ugly. Social play works though, and that's what we call it these days. Thanks to whoever came up with that!

Stop distracting wireless led blinking

Osma — Wed, 21 May 2008 17:34:00 +0300

While there's lots to like about my current laptop, one thing that had been quite annoying was the wireless indicator led. For a long time, it didn't function at all, because iwl4965 didn't contain support for driving it. Recently (effectively starting with Fedora 9 for me), that support came in, but now it's not annoying because you can't tell whether wireless is enabled, rather because the led is blinking all the time, which is a distraction.

I liked the behavior of my old laptop's ipw2200 much better: blink while searching/associating with a network, and then stay on constantly. Happily, Erich Schubert just pointed out how to fix the iwl4965 blinking behavior. That script is (I think) for Debian/Ubuntu, and a slightly different kind is needed on Fedora. I'm not sure this is the best way to go about it, but at least it works for me: put the following in /etc/NetworkManager/dispatcher.d/iwl-no-blink

#!/bin/sh if [ "$0" = "wlan0" ]; then for dir in /sys/class/leds/iwl-phy*X; do echo none > $dir/trigger done fi

ION 08 wrap-up

Osma — Fri, 16 May 2008 03:11:00 +0300

Finished my own lecture, Reinventing Habbo a couple of hours ago to a half-full audience. Flew through 174 slides in a bit over 30 minutes. Update: I have now made a shorter version of the deck that works better as a reference, below.

Reinventing Habbo - Upload a doc

The follow-up round table probably was titled very badly, since we got nearly no attendees. Interesting discussion though, small groups get deeper in.

Apart from that, I have a collection of 40-50 business cards from interesting new acquaintances, and several very good discussions I'm looking forward to continuing and/or applying to what we're doing. All in all, good experience. Still, I'm looking forward to getting back home too.

2nd update: Adam Martin has coverage of the event, including notes from one of the panels I was on.

Last-minute preparations for ION 2008, and issues with Shockwave

Osma — Sun, 11 May 2008 15:31:00 +0300

I'm leaving early tomorrow morning to Seattle for the ION 2008 online games conference. I'm looking forward to this trip quite a lot - I haven't been to this conference before, but I've heard good things about it. It's also the first time I'm doing a one-hour presentation in an American conference in front of an audience that I expect to be pretty much experts in the same field as I am. That's for sure going to be interesting, I hope the turnout is good. Been working to finalize the content of the session for a while, but I'm still to start compiling the slide deck; that's going to be a job in itself since I certainly don't want to be left second compared to Sulka's 160-slide rapid-fire session in GDC San Francisco that I now notice he never blogged about.

Somehow I ended up scheduled to appear in two panels in addition to that presentation and the round-table discussion that follows it. Last I checked, that could mean I have the most appearances in the whole agenda; I certainly didn't intend that when I signed up! :) All in all, that, half a dozen other sessions I'm really looking forward to hearing, and a few meetings in between will certainly keep me busy for the week.

And at the same time, I'm wishing I could cancel the whole trip, because of other work challenges.

With now 6 weeks after Shockwave 11's global rollout, most active Habbo users have probably encountered the automatic update mechanism which should look for a new version every 30 days. Unfortunately, Shockwave 11 has not been a great experience for us. Three distinct classes of problems are known to us:

1. The installer/auto-updater seems to function incorrectly in some operating system and browser combinations. In particular, errors have been seen on Windows Vista as well as Japanese language Windows, but issues have been reported also by Windows XP users in other countries. Adobe says it's due to third party Xtras, not because of Shockwave itself, but that's all the same to us because we can't control what Xtras might be on the users computer due to other games requiring them (Habbo doesn't require any).

2. Shockwave 11 changed some core functionalities in a backwards incompatible fashion in order to correct several long-standing bugs which most developers have years ago started considering as platform features. Eg, text handling details cause rendering mishaps with old content.

3. In order to provide compatibility with content authored for SW 10 or earlier, the player builds in a Shockwave 10 compatibility component stack due to large-scale developer demand (us included). However, this stack was put in at a late stage in the development cycle, and turns out has its own installation-related problems. In short, many new installations or upgrades to Shockwave 11 fail to execute content compatible with older versions on the first attempt, because downloading the compatibility components fails.

Each of these three groups contains multiple smaller issues, meaning it's been very difficult to identify clear root causes. At this point in time, perhaps the only highlight of the release is native support for Intel Macs, thus removing the requirement to use Rosetta mode for accessing Habbo on a Mac.

We have been in constant communication with Adobe to resolve these issues, but little concrete, measurable progress has been made. Instead, all through April and the beginning of May we've seen increasing reports of access problems and shorter sessions in the game, which we take to mean active users are growing tired of Shockwave problems. This is of course a reason for alarm.

To avoid the access issues caused by problem 3, we're now looking to switch to native Shockwave 11 mode, but in a way which does not force an immediate update for users still happy with their SW 10 installation, since Shockwave 11 will be less common than 10 for some time to come. Having to maintain compatibility for both versions 10 and 11 will likely cause some user-visible bugs, eg graphics glitches of the kind we would typically want to avoid. However, currently it looks like it's going to be an overall better experience to introduce a few glitches like that to everyone than to have a significant and growing segment of our most loyal userbase unable to access Habbo.

Of course the team can deliver on this kind of changes on their own, and even with a few other colleagues traveling next week as well, I know I'm not really going to be needed there to find a solution for this issue. Still, it's never fun to leave the action behind on a critical moment like this looks to be.

RIght move, MySQL

Osma — Sat, 10 May 2008 16:14:00 +0300

Again a week late, but hey, I only need to keep up with this stuff, not comment on it all the time. MySQL changed their minds and turns out the core server will continue to be open source, allowing customers to depend on being able to inspect it if required, extend on any bit as needed, and most importantly, get the benefits of a large community using and testing all features. Thanks for that. I just hope you're going to be consistent about this, for precisely the reason that as a MySQL Enterprise customer, I don't pay you to deliver bits that haven't received that community testing, but to rapidly fix problems if they exist despite that exposure.

It was interesting to hear Monty Widenius comment about it in this week's Open Tuesday event, and I also got to talk to him about attending a MySQL Users session in Helsinki next time I or someone else (anyone? anyone? Bueller?) manage to organize one. Would be nice to hear about the upcoming storage engines straight from the horse's mouth - Monty's Maria effort has certainly been less covered than the Falcon engine I have also commented on, and I can't say to know anything about it myself.

MySQL Users Conference followup and MySQL's business model

Osma — Tue, 22 Apr 2008 18:20:00 +0300

Last week saw MySQL User Conference 2008 in Santa Clara, but I was not able to make time for it this year either. However, in the wake of Sun's acquisition of MySQL, it was very interesting to follow what was going on. A few things that caught my attention:

MySQL 5.1 is nearing General Availability and an interesting storage engine plugin ecosystem starts to emerge. It's this latter, but related event that I see as the first real sign of validation for MySQL's long-ago chosen path of pluggable storage systems instead of focused effort on making one good general-use engine.

Oracle/Innobase announced InnoDB Plugin for MySQL 5.1, which much-awaited features which promise a great deal of help for daily management headaches. More than that, InnoDB Plugin's release under GPL lifts quite a lot of the concern I'm sure many users like us have had about the future viability of InnoDB as MySQL storage engine.

A couple of data warehousing solutions are launched, also based on MySQL 5.1 -- Infobright is one I've already researched somewhat (looks very interesting, as soon as a few current limitations are lifted), Kickfire I know nothing about right now but would love to learn more of.

There's a huge amount of coverage graciously provided by Baron Schwartz that I have yet to fully browse through.

A few remarks by Mårten Mickos regarding MySQL's business model seem to have kicked up a bit of a sandstorm. I don't really understand why; I read these to just verify that the direction MySQL took last year is to continue this year as well. I don't see any major changes here regarding the licensing structure, software availability, or support models. Frankly, it seems like yet another case of Slashdot readers not reading, let alone understanding, what they're protesting against, and press following up on the noise.

I do understand the critique made against MySQL's chosen model, though. In fact, I went on record last September to say that I understand that critique. I still see the same issues here. I believe we represent a fairly common profile of a MySQL Enterprise customer in that what we want from it is not the bleeding-edge functionality but a stable, well-tested product that we can expect to get help for if something does go wrong. We don't see great value in having access to a version of software that isn't generally available to "less advanced" or more adventurous users for free in a community version. In fact, we see it as a negative that such functionality exists, because it hasn't received the community testing, feedback and improvements that makes great open source software as good as it is. While new functionality is interesting, and we're trying to spend time getting familiar with new stuff in order to use it in production later, it simply isn't prudent to put business-critical data in a system that hasn't received real-world testing by as large a community as possible (unless you have no other alternative, and then you takes your chances).

Yet it seems to me that this is essentially what Sun/MySQL continue to propose for the Enterprise customers by delivering "value add" functionality in a special version of the server or plugins to it, possibly in a closed-source form that further reduces transparency and introduces risk. Mårten, I'd prefer it to be otherwise. How can I help you change your mind about this?