Fishpool - Comments

First thoughts about Balancion - Osma

Osma — Thu, 04 Feb 2010 15:59:01 +0200

Lähetin pari päivää sitten melko kiukkuista asiakaspalautetta OP:lle Balancionin kieltämisestä, ja tänään tuli hieman erilainen viesti takaisin: lyhennettynä "OP ei voi taata palvelun turvallisuutta, käyttö on asiakkaan omalla vastuulla". Tähän voisi sanoa "no shit, Sherlock" - mutta ainakaan he eivät varsinaisesti KIELLÄ käyttöä.

First thoughts about Balancion - Osma

Osma — Mon, 01 Feb 2010 15:14:43 +0200

So, just before the weekend OP decided they want to ban Balancion referring to security concerns. To hell with them - sure it's insecure to log in to a bank using a custom browser designed to take over the web session as soon as I'm authenticated, and I'd much rather not do that. However, that's not an option unless banks offer a decent read-only API and OAuth-style authentication for a client such as Balancion's to grab the transaction information without resorting to screen scraping.

So, OP, don't talk the talk unless you're prepared to walk the walk. I'm waiting.

First thoughts about Balancion - Jussi Muurikainen

Jussi Muurikainen — Sun, 17 Jan 2010 17:26:22 +0200

Hi Osma,

Very thorough pre-analysis about Balancion.

I´m glad to inform that Your suggested "next steps" are pretty much what we are planning with Idean almost as we speak ;)

Regards,

Jussi

Jussi Muurikainen
founder & CEO
Balancion Ltd.

PS. for those who want to know more in English:

http://www.finnfacts.com/english/ne...

http://www.arcticstartup.com/2009/0...

The balance of great products and rapid evolution - Sampo Karjalainen

Sampo Karjalainen — Tue, 05 Jan 2010 20:36:46 +0200

Good challenge. Some thoughts from my fluish mind:

Great design is about visualizing concrete solutions that truly serve people's goals, needs and wants.

People use products to get 'a job' done. The job can be built of...
* concrete end goals like "I want to build and share a photo slide show"
* experience goals like "I want my photos to look really good"
* or higher level life goals like: "I want to be respected by other photographers."

In one product, the individual features, UIs or layouts are all there to build a flow that gets that job done. A job done – and you are ready to pay for it in one form or another, and probably recommend to your friends, too.

The challenge with granularity and design is that to satisfy those goals, you need to see the product, the users and the context as a whole. Putting up a photo upload functionality for the user is a start. But to get your slide show up on your friend's computer, and get praise from him, may require a sequence of seamlessly integrated features of uploading, scaling, ordering, previewing, sharing, commenting, etc. Before the chain is complete, all we have is some random meaningless features (for this particular job).

Once you have the flow designed, it's quite easy to design the individual features one by one and take them further incrementally and iteratively. But to serve a new need requires a holistic view of the whole design.

Some scaling observations on Infobright - david!lutz

david!lutz — Thu, 26 Nov 2009 00:57:10 +0200

Osma,

Nice observation, test and summary. Thanks!

David Lutz
Director, Technical Sales Consulting
Infobright

[ping] MySQL - could we please move on already? - uberVU - social comments

uberVU - social comments — Wed, 11 Nov 2009 22:37:37 +0200

Social comments and analytics for this post

This post was mentioned on Twitter by osma: MySQL - could we please move on already? http://bit.ly/X73mx...

MySQL - could we please move on already? - Kennu

Kennu — Wed, 11 Nov 2009 15:23:02 +0200

Better to just jump over to NoSQL already... Then we can forget all that MySQL stuff ;-)

[ping] Habbo, soon ten years old, still the leader - Habbo Herald

Habbo Herald — Sat, 07 Nov 2009 11:03:31 +0200

Habbo still #1

Habbo proves it is still the best when the results for new user registrations for the 3rd quarter of 2009 show up and tell us how popular Habbo really is. ......

Habbo, soon ten years old, still the leader - jamcs

jamcs — Sun, 01 Nov 2009 17:51:23 +0200

Please re-launch it to the iPhone/iPod touch for a limited time as well as other phones<3

Creating your very own Habbo site - Lab-Rat

Lab-Rat — Sun, 01 Nov 2009 16:20:54 +0200

Nice, for a second there I thought this was gonna reference Retros and was like o.O, Might try this :)

Habbo, soon ten years old, still the leader - Ryan1806

Ryan1806 — Sun, 01 Nov 2009 14:40:57 +0200

Relaunch the disco!

Habbo, soon ten years old, still the leader - Lab-Rat

Lab-Rat — Sun, 01 Nov 2009 14:40:36 +0200

Wouldn't mind playing the game that started off Habbo, and congrats to Habbo for still being the leader :)

Habbo, soon ten years old, still the leader - Laurensh

Laurensh — Sun, 01 Nov 2009 14:22:51 +0200

Please relaunch the Disco. That would be superawesome best thing EVER.
Thnx.

Keep it going :D

A peek under the hood in Infobright 3.2 storage engine - Osma

Osma — Fri, 25 Sep 2009 14:21:38 +0300

Some early notes on query performance between Infobright 3.1.1 and 3.2:

LEFT JOINs are significantly faster (many times faster for large data sets)
queries which were impossible to execute now work (such as the major new feature, aggregate functions over calculated values)
a regression in the joiner makes joins really slow if the columns being joined don't have the exact same data type. Watch out for "Starting joiner loop"
a trick which worked around 3.1.1 limitations, deeply nested subqueries, seems to sometimes work slower now. However, it's less likely to be needed now that more things can be expressed without the use of subqueries

A peek under the hood in Infobright 3.2 storage engine - Osma

Osma — Fri, 25 Sep 2009 14:07:12 +0300

I don't think the storage engine would build as a plugin, I believe they've had to change parts of the upper MySQL layers to inject the columnar execution model into the system. The server does not include InnoDB, and I have not tried installing InnoDB plugin nor building a server with InnoDB included.

Even if the above were possible, I would still say you're better of putting the analytics-oriented tables into their own instance or preferably an entirely separate machine from the InnoDB server, in particular if you're concerned enough about performance to run the Percona builds. I would recommend this even if you were using MyISAM as the table engine for the analytics tables. It's really not the same database system, even though the same clients work, because the workload and with it, the correct server tuning is entirely different.

A peek under the hood in Infobright 3.2 storage engine - jon

jon — Fri, 25 Sep 2009 04:54:57 +0300

looks interesting - just about to try infobright myself. we currently run the percona built mysql binaries - is it possible to use infobright as a plugin? is innodb available on the infobright version?

Reflections on Nokia Maemo - Osma

Osma — Mon, 31 Aug 2009 15:53:40 +0300

Huh. Just take a look at this presentation, listen to the audio track from minutes 9:00 to 11:00 (thats slide 5 onwards, approximately), and come back here to tell me what was the crucial point I didn't get. Oh, and provide references next time. Thanks.

Reflections on Nokia Maemo - Mikko

Mikko — Mon, 31 Aug 2009 08:59:58 +0300

You're not getting a crucial point. Nobody is *switching* to anything. Qt is being introduced on S60 and Maemo but AVKON and GTK will continue to work. So your point "won't work on the next device" is completely false.

Why mobile computers are a bad idea - Osma

Osma — Sat, 29 Aug 2009 22:51:29 +0300

Have you noticed how in the last couple of years, the word laptop has practically speaking replaced computer in the everyday consumer language? Now we call only desktop machines computers, and yet for some reason one should also fit into a pocket. Still, I'm afraid I haven't figured out what would be a good substitute term. PDA communicated something relevant, but it's an acronym out of something really long and boring -- and eventually was understood to mean only a very limited set of early functionality.

Why mobile computers are a bad idea - Joe Ludwig

Joe Ludwig — Sat, 29 Aug 2009 18:39:22 +0300

Ah, so it's the word "computer" you don't want in your pocket. I guess I could have said, "I want something programmable with storage that can compute in my pocket, preferably hooked up to some sort of head-mounted display." :)

You have a good point though. Unfortunately I haven't found my iPod Touch or my Android phone to be much more reliable than my desktop windows PC.