[ABOUT LEAKS] What I can tell from a simple mistake.

A while ago, Microsoft accidently leaked Tulalip. Some social network experiment that could be

• a MVP smoke test (see http://en.wikipedia.org/wiki/Minimum_viable_product)
• an unfinished product 

If it was a smoke test, it was both cleverly planned and very silly at the same time.
Why? Because the timing would have been too obvious (15th of July, shortly after the Google+ launch).

source: searchenginewatch.com

Well Microsoft was already working on it since a while as a Google search reveals:

(the search results have already been removed in the meanwhile)

You can see several instances of the same application.
The Google cache shows me that the site was already crawled on the 27th of June 2011

The biggest rookie mistake was to forget to exclude the site from being crawled by using a simple robots.txt
(sorry for the hindsight, such things can happen, I know)
Another reference I found was the use of a free load-test (see here http://loadimpact.com/result/tulalip-scratch.cloudapp.net/-6234fd9c421156d2dab04dcadedcddbe ) performed on the 17th of June 2011 for the domain tulalip-scratch.cloudapp.net

And now the pages just show plain and simple a message from (or on behalf of) the dev. team:

Furthermore there is one single very interesting fact:

The domain share-film.com is the host for a warez page and.. surprisingly also the host of an older version of Microsofts social net.
Hm. Funny... the image-preview shows a different design and the original name of the social network...
Makes sense after the leak was discovered on www.socl.com

ser.sharefilm.com has also some different footprints here on the web...

Ok ok, whatever... all this isn't so important.
More important is, what we can learn about Microsoft's way of testing and developing products.

Tulalip/Socl is basically a smaller project created by a small dev. group... creative, but not that experienced. I'd say they are mostly fresh from University or school.
If it were a well-funded project, it wouldn't have leaked in that way. They'd have security experts having an eye on the project... making sure that it isn't indexed by the Googlebot.

Furthermore, they'd consider a different way of testing their software than to just use a free load-test somewhere on the web that compromises the security of your project by posting the URL and the results on the web.

Microsoft would have it's own testing software for that...

Too many rookie-mistakes for a well experienced dev-team or a project with a big budget.

Microsoft has most likely multiple small groups working on different projects, all hosting on cloudapp.net (I guess some kind of AppEngine made by Microsoft, hosting .net projects)

There are two subdomains I found on cloudapp.net, referring to Socl/Tulalip:

• tulalip.cloudapp.net
• tulalip-scratch.cloudapp.net

I guess one was the testing version and the other one was the dev-version.
So, if someone interested in leaks might want to hunt, the best way to get started is to analyze habits, best practices and procedures of a company or developers.
In this case I guess the hunt wouldn't make much sense as the group seems to be not really "into" or tied to company internal best practices... but, if I'd want to hunt for THEM, I'd try to find every possible subdomain on cloudapp.net.

It is highly unlikely that they have simply shut down the application. It is more likely that they moved/renamed the instance somehow (this would cause way less trouble than to move it to a completely different server)

My best guess is, that it is still hosted somewhere on cloudapp.net... I just have to wait for Google to index it again. (if they haven't learned from their mistakes... they are rookies)
The rename from socl to tulalip (even if they had the domain socl.com) also tells me, that it is not a primary focus of Microsoft. (no money, no lawyers)

The reason why they had to rename it can be found here:

http://socl.tntlab.org/index.php?title=The_SOCL_Project:_Social_networking_for_Online_Community-based_Learning

Yep.. Ooops. (dated 4th of July 2011)

Well... now I know where Microsoft might have leaks... cloudapp.net.... I'm watching you.

And my other guess: they might set up a second serverfarm for internal development too... hmm.. we'll see.

BTW... Warez site? WTF? I hope the guy who upped the early build of socl has a very good explanation for that!
I like the design and I hope they don't slash the project just because of some... stupid mistakes.
I probably won't use it... but hey, it looks great :)

A while ago, Microsoft accidently leaked Tulalip. Some social network experiment that could be

• a MVP smoke test (see http://en.wikipedia.org/wiki/Minimum_viable_product)
• an unfinished product 

If it was a smoke test, it was both cleverly planned and very silly at the same time.
Why? Because the timing would have been too obvious (15th of July, shortly after the Google+ launch).

source: searchenginewatch.com

Well Microsoft was already working on it since a while as a Google search reveals:

(the search results have already been removed in the meanwhile)

You can see several instances of the same application.
The Google cache shows me that the site was already crawled on the 27th of June 2011

The biggest rookie mistake was to forget to exclude the site from being crawled by using a simple robots.txt
(sorry for the hindsight, such things can happen, I know)
Another reference I found was the use of a free load-test (see here http://loadimpact.com/result/tulalip-scratch.cloudapp.net/-6234fd9c421156d2dab04dcadedcddbe ) performed on the 17th of June 2011 for the domain tulalip-scratch.cloudapp.net

And now the pages just show plain and simple a message from (or on behalf of) the dev. team:

Furthermore there is one single very interesting fact:

The domain share-film.com is the host for a warez page and.. surprisingly also the host of an older version of Microsofts social net.
Hm. Funny... the image-preview shows a different design and the original name of the social network...
Makes sense after the leak was discovered on www.socl.com

ser.sharefilm.com has also some different footprints here on the web...

Ok ok, whatever... all this isn't so important.
More important is, what we can learn about Microsoft's way of testing and developing products.

Tulalip/Socl is basically a smaller project created by a small dev. group... creative, but not that experienced. I'd say they are mostly fresh from University or school.
If it were a well-funded project, it wouldn't have leaked in that way. They'd have security experts having an eye on the project... making sure that it isn't indexed by the Googlebot.

Furthermore, they'd consider a different way of testing their software than to just use a free load-test somewhere on the web that compromises the security of your project by posting the URL and the results on the web.

Microsoft would have it's own testing software for that...

Too many rookie-mistakes for a well experienced dev-team or a project with a big budget.

Microsoft has most likely multiple small groups working on different projects, all hosting on cloudapp.net (I guess some kind of AppEngine made by Microsoft, hosting .net projects)

There are two subdomains I found on cloudapp.net, referring to Socl/Tulalip:

• tulalip.cloudapp.net
• tulalip-scratch.cloudapp.net

I guess one was the testing version and the other one was the dev-version.
So, if someone interested in leaks might want to hunt, the best way to get started is to analyze habits, best practices and procedures of a company or developers.
In this case I guess the hunt wouldn't make much sense as the group seems to be not really "into" or tied to company internal best practices... but, if I'd want to hunt for THEM, I'd try to find every possible subdomain on cloudapp.net.

It is highly unlikely that they have simply shut down the application. It is more likely that they moved/renamed the instance somehow (this would cause way less trouble than to move it to a completely different server)

My best guess is, that it is still hosted somewhere on cloudapp.net... I just have to wait for Google to index it again. (if they haven't learned from their mistakes... they are rookies)
The rename from socl to tulalip (even if they had the domain socl.com) also tells me, that it is not a primary focus of Microsoft. (no money, no lawyers)

The reason why they had to rename it can be found here:

http://socl.tntlab.org/index.php?title=The_SOCL_Project:_Social_networking_for_Online_Community-based_Learning

Yep.. Ooops. (dated 4th of July 2011)

Well... now I know where Microsoft might have leaks... cloudapp.net.... I'm watching you.

And my other guess: they might set up a second serverfarm for internal development too... hmm.. we'll see.

BTW... Warez site? WTF? I hope the guy who upped the early build of socl has a very good explanation for that!
I like the design and I hope they don't slash the project just because of some... stupid mistakes.
I probably won't use it... but hey, it looks great :)

[ABOUT LEAKS] What I can tell from a simple mistake.

Microsoft leaks "accidently" a social experiment.

You can run, but you can't hide.

interesting.... and digging deeper.

And if you haven't realized yet... this is the subdomain of a warez-site?! WTF?!

It is hosted on cloudapp.net... so there is a chance they just renamed the app-instance... The hunt is on!

http://fusible.com/2011/07/new-microsoft-social-search-service-called-tulalip-revealed-on-socl-com/

You can run, but you can't hide.

interesting.... and digging deeper.

And if you haven't realized yet... this is the subdomain of a warez-site?! WTF?!

It is hosted on cloudapp.net... so there is a chance they just renamed the app-instance... The hunt is on!

http://fusible.com/2011/07/new-microsoft-social-search-service-called-tulalip-revealed-on-socl-com/

Microsoft leaks "accidently" a social experiment.

One or two things about keyboards. [Part 1]

Yeah, I gotta admit: I am a programmer. Therefore I've quite a fetish when it comes to keyboards.

I simply can't work with a crappy keyboard, so I went through quite some lengths to find a good one.

First: No, I don't really hate a numpad... but if you have a limited workspace, like me, you will go nuts at least the 10000th time, you bump with your mouse on the way too big keyboard.

So you start looking for alternatives.
You might think: Oh, let's check for a wireless solution... probably something from a well known brand... not too expensive... and there you go with the Microsoft Arc Keyboard

Microsoft Arc Keyboard

First things first: It comes with a reaaaaally nice cloth thing to store it somewhere... kinda protective covering. I guess most people throw it away, but it was just too nice to do that, so I kept it. It is not a Bluetooth keyboard, so you get this little transceiver that is cleverly stored under the keyboard itself using a magnet. This can be a pro or con. If you still own floppy disks, don't put this keyboard on top of them.

This thing runs on AAA juice... not quite the money friendliest power-solution (at least here in Austria it isn't), but I haven't had to replace the batteries yet. (not bad, not bad)
So, here are the con's: It is one of those super-shiny-glossy-accessories that you will love at first and damn later, because it attracts fingerprints and smudge like nothing else... and a keyboard is way more difficult to clean than a touchscreen. Second, very uncomfortable fact is, that the keyboard is too flat to be comfortable for typing... It is super-portable, super stable on the desk, but simply too flat on the edges... next problem is the way the keys feel while typing.. I always hat the Impression there was too much friction while I typed. It felt like the keys would start to wear off quickly... not so pleasant. And here is the biggest downside: the d-pad or whatever that should be.

Nightmare.

It is almost impossible to use. I guess it was designed for powerpoint presentations or something like that, because it is as userfriendly as the left and right click buttons on the Acer Aspire One A150.
Drivers are sometimes an issue too, so you might check if you plugged in the transceiver into a different USB port than last time, because sometimes the driver does not load and the keyboard does not work just because of that.
Overall: Nice try... It is definitely a nice keyboard, but not for programmers or people who write a lot... or that use the arrow keys to navigate. I'll still use it for XBMC... But coding? No way.
Uh yeah, before I forget, the price... It is about 33$ on Amazon.com or 33€ on Amazon.de

Second try: The infamous Apple Wireless Keyboard

Yeah I know... "Booo, you bought from the enemy!". I just needed a Bluetooth keyboard that is 100% working with my Samsung Galaxy Tab :-P
Of course it is fancy packaged and smells like Apple things smell... Pairing was easy and even batteries were included.
Yeah it has arrow keys... big plus. But the arrangement of the keys is different & "<" and "^" keys are exchanged (default mapping)... ok, I can live with that. It has a similar typing experience as the Microsoft keyboard... slightly better though, but not so much better that it would justify the 66€ I paid for it... and most annoying thing: no "DEL" key o.O That really sucks. and I mean really really...
I haven't had much time to test the battery-life though, so I'll probably report another time about that :)

So much for the first part... but I'll add a little information another time.

Yeah, I gotta admit: I am a programmer. Therefore I've quite a fetish when it comes to keyboards.

I simply can't work with a crappy keyboard, so I went through quite some lengths to find a good one.

First: No, I don't really hate a numpad... but if you have a limited workspace, like me, you will go nuts at least the 10000th time, you bump with your mouse on the way too big keyboard.

So you start looking for alternatives.
You might think: Oh, let's check for a wireless solution... probably something from a well known brand... not too expensive... and there you go with the Microsoft Arc Keyboard

Microsoft Arc Keyboard

First things first: It comes with a reaaaaally nice cloth thing to store it somewhere... kinda protective covering. I guess most people throw it away, but it was just too nice to do that, so I kept it. It is not a Bluetooth keyboard, so you get this little transceiver that is cleverly stored under the keyboard itself using a magnet. This can be a pro or con. If you still own floppy disks, don't put this keyboard on top of them.

This thing runs on AAA juice... not quite the money friendliest power-solution (at least here in Austria it isn't), but I haven't had to replace the batteries yet. (not bad, not bad)
So, here are the con's: It is one of those super-shiny-glossy-accessories that you will love at first and damn later, because it attracts fingerprints and smudge like nothing else... and a keyboard is way more difficult to clean than a touchscreen. Second, very uncomfortable fact is, that the keyboard is too flat to be comfortable for typing... It is super-portable, super stable on the desk, but simply too flat on the edges... next problem is the way the keys feel while typing.. I always hat the Impression there was too much friction while I typed. It felt like the keys would start to wear off quickly... not so pleasant. And here is the biggest downside: the d-pad or whatever that should be.

Nightmare.

It is almost impossible to use. I guess it was designed for powerpoint presentations or something like that, because it is as userfriendly as the left and right click buttons on the Acer Aspire One A150.
Drivers are sometimes an issue too, so you might check if you plugged in the transceiver into a different USB port than last time, because sometimes the driver does not load and the keyboard does not work just because of that.
Overall: Nice try... It is definitely a nice keyboard, but not for programmers or people who write a lot... or that use the arrow keys to navigate. I'll still use it for XBMC... But coding? No way.
Uh yeah, before I forget, the price... It is about 33$ on Amazon.com or 33€ on Amazon.de

Second try: The infamous Apple Wireless Keyboard

Yeah I know... "Booo, you bought from the enemy!". I just needed a Bluetooth keyboard that is 100% working with my Samsung Galaxy Tab :-P
Of course it is fancy packaged and smells like Apple things smell... Pairing was easy and even batteries were included.
Yeah it has arrow keys... big plus. But the arrangement of the keys is different & "<" and "^" keys are exchanged (default mapping)... ok, I can live with that. It has a similar typing experience as the Microsoft keyboard... slightly better though, but not so much better that it would justify the 66€ I paid for it... and most annoying thing: no "DEL" key o.O That really sucks. and I mean really really...
I haven't had much time to test the battery-life though, so I'll probably report another time about that :)

So much for the first part... but I'll add a little information another time.

One or two things about keyboards. [Part 1]