TurnKey Linux Virtual Appliance Library

TurnKey Appliance Development Contest: An Open Source Summer Bonanza!

Over the last few months donations have been trickling in and gradually piling up. Since there's a limit to how much beer we can reasonably drink we've been brainstorming ideas for using that money to help the project.

One of the ideas we liked the most was to try and sponsor an experimental contest which would hopefully stimulate community TKLPatch development, strengthen the community and teach as many people as possible the skills needed to customize existing TurnKey appliances and create new ones. It's not rocket science and you don't need to be a programmer or systems expert. Anyone willing to learn a few basic Linux skills can have fun doing it. If you want we'll even teach you how! (details below).

We're thinking of running the contest for the next 8 weeks. Then allow another 2 weeks to summarize the results and let the community help us decide who should win (e.g., a survey on the web site).

Prizes

Winners and all honorable mentions will be celebrated in obligatory blog posts, and forever immortalized in the TurnKey hall of fame. They'll also receive full bragging rights, the undying gratitude of billions of TurnKey users, and some of our excess beer money on PayPal, which they can claim for themselves or donate to a charity of their choice.

  1. First prize: $1,500 $100,000 + pony
  2. Second prize: $800
  3. Third prize: $100

Many thanks to everyone who donated to the project! We're hoping this will put your money to good use. If you like the idea and want to increase the contest pot size, feel free to donate now and ask us to dedicate the money to that.

Help us expand the next release

We're hoping to get lots of high-quality submissions, and if we do, the timing will be just right.

As many of you already know we're in the middle of a development cycle for the new Lucid and Lenny based beta appliances and we have our hands full upgrading the existing crop of appliances and introducing new features such as backup and migration. Unfortunately that means this time around we don't have the resources to expand the appliance library horizontally and add a significant number of new appliances from scratch.

Unless... well, that's where our heros come in. Adding a new appliance to the library from a TKLPatch still requires work, but it's definitely easier than creating an appliance from scratch and that means we could get more of them done.

In other words, any high-quality TKLPatch submissions we receive from the community in the next couple of months will have a very good chance of being formally adopted into the library in time for the next release.

After that the appliance library will be frozen again for a while until the next release batch.

List of ideas

In no particular order here's a list of ideas for appliances that have been sitting in our todo list for much too long and probably won't make it into the next release without the community's help:

  • By function (components not yet determined): terminal server replacement (I.e., remote desktops for thin-clients), ASP.NET replacement, web filtering proxy (e.g. privacy, ad-blocking, malware protection, content filtering), plug-in e-mail filter (e.g., spam/malware), unified threat management, load balancing reverse proxies.
  • eCommerce: Magento, VirtueMart, PrestaShop, Zen Cart, osCommerce, UberCart.
  • Content management: Alfresco, SilverStripe, Plone, Knowledge Tree, DSpace, Apache Roller, LifeRay.
  • Messaging: DimDim, Asterisk, OpenFire, Mumble, Vanilla forum, StatusNet, Zarafa, Scalix, RoundCube, OpenEMM
  • Business: SugarCRM, vTiger CRM, Open HRM, Apache OFBiz, GLPI
  • Monitoring: Nagios, Cacti, ZenOSS, Hyperic, Zabbix, OpenNMS
  • Infrastructure: OpenLDAP, Radius
  • Development frameworks: Zope, TurboGears, jBos
  • Management: eBox, ISPConfig
  • Data integration: Jasper BI, Pentaho, SnapLogic
  • Backup: Amanda, Bacula
  • IDS: Snort + Aanval, Snorby
  • Virtualization: Proxmox VE, Xen-DTC, oVirt, enomaly, eucalyptus
  • Misc: Apache Solr, iFolder, IceCast, OpenVPN ALS

If you have your own ideas for appliances you think would make good additions to the TurnKey library, don't worry if they're not on the list. All contributions are welcome. Work on whatever interests you.

Sign up for a live training session

Creating a TKLPatch isn't hard. Read the documentation and still not sure where to begin? No problem. If there's interest, we'll be giving free live training sessions on TurnKey's IRC channel showing how to build an example TKLPatch step by step and answering any questions. Click here to sign up and we'll send you an e-mail with the time and date of the session.

UPDATE: The session is scheduled for July 25 2010, 17:00 UTC

A few guidelines

  • You can improve an existing appliance or create a new appliance by patching any appliance in the TurnKey library, including the new Ubuntu Lucid and Debian Lenny TurnKey Core betas.

  • If you're creating a new appliance, we recommend patching the new beta of TurnKey Core based on Ubuntu 10.04 Lucid. This is the latest version of Ubuntu so it has the newest packages in its software repositories.

  • It's preferable to install software through the package management system rather than directly from an upstream tarball. It's usually much easier to install and update software this way.

    Unfortunately, sometimes this won't be an option because there's a lot of excellent open source software that isn't in Ubuntu's or Debian's package repositories. In these cases try checking if the software you're looking for is at least available as a Debian package (*.deb).

    If not, that's OK. An appliance doesn't have to be perfect to be useful.

  • In case it isn't obvious, if you include software from outside the official package repositories, make sure it's available under an open source license (e.g., GPL, BSD, etc.). Free as in free beer is not enough. Software in official TurnKey appliances must also be free as in liberty.

  • Publish results as soon as you have them on the forum and/or development wiki. In general, credit for a result goes to the first person who publishes. This doesn't even have to be a finished TKLPatch, though naturally finished, high-quality submissions count moure than partial results.

  • To avoid duplicated effort, check the development wiki before you start working on a new TKLPatch. Maybe it's already been submitted. But if you think you can make something better, don't let that stop you!

  • You can work alone, or collaborate as part of a group. Your choice. There's no need to register. If you're part of a group, just document who should take credit (e.g., Hans Solo, Star Wars group).

  • Results will be evaluated based on a quality and quantity of all submissions, including any integration notes, TKLPatches, or documentation submitted. At the end we'll summarize the results of all participants and set up community surveys to help us decide who should win.

  • Have fun!

Don't forget to sign up for the live training session if you're interested. Any questions?

You can get future posts delivered by email or good old-fashioned RSS.
TurnKey also has a presence on Google+, Twitter and Facebook.

Comments

What?!? No Pony?

I was excited about this contest until I read the list of prizes and saw that it did not include a pony.

Alon Swartz's picture

Pony's are available by request

Pony's are available by request, but pink pony's cost more - and are not in stock - sorry.

Maybe next time...

Liraz Siri's picture

Donate a pony, and we'll add it to the pot

If someone donates a pony and asks that we add it to the contest pot, we can make it happen. You can contact me privately to discuss shipping and handling.

If you don't have a pony, consider donating a pony equivalent (e.g., a pony bond).

We also accept cash, precious gems, and special favors.

All this talk of pots and ponies is making me thirsty

Amusingly (for varying values of ~) pot and pony are both measures of beer in Australia.


Liraz Siri's picture

G'day mate

No wonder you have so many bogans down under. I figure you'd get a nice mob of people rotten full with a pony keg of grog. (Australian slang is the best)

Is it possible to participate with more than one appliance?

This development contest is very interesting for me, now that I have more experience with Linux maybe I'll could finally make the Alfresco appliance I planned a year ago. But there's also a lot of TKL appliance I would like to see, and the contest motivates me to participate with 2 or 3 appliances at least. Is it possible to participate with more than one? 

P.S.: What's that about pink ponys? I don't get the joke at this side of the world :S

Liraz Siri's picture

The more the merrier!

Glad to have you on board Adrian! Contributions will be evaluated both on quality and quantity. There's an extensive todo list of missing appliances so it wouldn't make sense to limit people/groups to just submitting one. We're encouraging everyone to submit as many contributions as they feel they're up for.

The more the merrier!

P.S: Pink ponies probably infected Alon's mind while he was working with Python/Django. It's a deadly mind virus. Be careful.

Liraz Siri's picture

Live training prerequisites

To participate in the live training session, you're going to need to setup a TKLPatch compatible build environment first. This will allow you to follow along and get your hands dirty with real TKLPatching examples.

The build environment is itself based on TurnKey so it's easy to get started:

  1. Download and deploy TurnKey Core to a virtual machine

    http://www.turnkeylinux.org/core

    If you don't already have virtualization software installed, we recommend VirtualBox. It's free:

    http://www.virtualbox.org/

    If this is your first time installing a TurnKey appliance into a virtual machine, don't worry it's very simple: download the VM build, import the OVF file in VirtualBox and set the virtual machine's network settings to "bridged" (if you still have trouble, follow the tutorial). Boot the machine, log into a shell (see below) and ping 8.8.8.8 to make sure your machine can access the Internet.

  2. Install tklpatch

    First you ened to log into a root shell. There are several ways to do that but one of the easiest is to use the web shell to get command line access from any browser. Point your web browser at:

    https://1.2.3.4:12320/

    Where 1.2.3.4 is replaced by the IP address of your virtual machine.

    Username: root
    Password: <blank>

    When you import an appliance into a virtual machine from the VM build the root password is empty by default until you set it.

    Now type in the following commands:

    apt-get update
    apt-get install tklpatch
    
  3. Download the TurnKey Core ISO into the virtual machine.

    While still logged into the shell type in the following commands:

    cd /root
    wget http://downloads.sourceforge.net/project/turnkeylinux/turnkey-core/2009.10-2-hardy-x86/turnkey-core-2009.10-2-hardy-x86.iso
    

You're done! You can test that tklpatch works by typing in the following commands:

cd /root
tklpatch-example helloworld
tklpatch turnkey-core-2009.10-2-hardy-x86.iso helloworld

Note: Ubuntu 10.04 (Lucid) use a different ISO filesystem standard which is incompatible with the existing Ubuntu 8.04 based appliances. That means to if you plan on creating TKLPatches based on the Lucid TurnKey Core beta you'll need to follow the above instructions with the Lucid-based TurnKey Core beta instead of the Hardy-based TurnKey Core.

Problem with steps 2 & 3

FYI,

tklpatch fails to download

Building dependency tree
Reading state information... Done
The following NEW packages will be installed:
tklpatch
0 upgraded, 1 newly installed, 0 to remove and 27 not upgraded.
Need to get 8410B of archives.
After this operation, 77.8kB of additional disk space will be used.
Get:1 http://archive.turnkeylinux.org hardy/universe tklpatch 0.93+1+ge3b62ec [8410B]
Fetched 1B in 0s (14B/s)

Failed to fetch http://archive.turnkeylinux.org/ubuntu/pool/universe/t/tklpatch/tklpatch... Size mismatch

E: Unable to fetch some archives, maybe run apt-get update or try with --fix-missing? 

 
Also,

wget http://www.turnkeylinux.org/download?file=turnkey-core-2009.10-2-hardy-x...

does not work....  Had to use  -->

wget http://downloads.sourceforge.net/project/turnkeylinux/turnkey-core/2009....

 

Thanks, look forward to the live training..


Alon Swartz's picture

apt-get install tklpatch - Fixed

Thanks for reporting both issues.

Regarding the error you received when trying to install tklpatch. It was caused by a conflict within the package archive between hardy and lucid. I've fixed the issue and you should be to install tklpatch now without issue.

See you at the tklpatch session.

Liraz Siri's picture

Fixed wget url

Woops. Thanks for catching this. I updated the wget url. What happened was that I wrote the instructions offline and forgot that we were using a javascript redirection for the download page.

Bottom line: never assume anything works. Always test.

Thanks guys :)

Alon & Liraz,

Thanks for the updates, all seems to work well under both hardy and lucid now:) 

Marc


Alon Swartz's picture

TKLPatch Live Training schedule

The session is scheduled for this coming Sunday, July 25 2010, 17:00 UTC and will be held in #turnkey on irc.freenode.net

Don't forget about the prerequisites.

See you there.

Patrick Binder's picture

why not use the Ubuntu Customization tool to create the live ISO

Their is a ubuntu customization tool that allows you to create custom Ubuntu version that have specific packages installes at installation. Basically you start the tool choose the linux ISO you want to start with and then modfiy the packages that you want installed during the installation. all we have to do is go ahead and create a repo list to modify that has only the packages for each distro. btw where is the beta turnkey dl at? only downside is it automatically makes you select a gui which we will later have to manually remove from the package list or set as disable by default so that it boots into rl2 instead of rl5 and allows for a user to run stratx to run the gui.


Alon Swartz's picture

TKLPatch vs. UCK

I assume you are referring to the Ubuntu Customization Kit (aka. UCK).

I won't go into a TKLPatch vs. UCK discussion, as I have never used UCK myself, just scanned the docs. What I will say (AFAIK) is that UCK was designed with a certain use case in mind (eg. customizing the ubuntu-desktop for localization, package changes and minor tweaks) to produce a Live CD.

OTOH, TKLPatch was designed to make it easy to customize Turnkey images with a very powerful interface, and produces a TKLPatch (which in turn can be used to produce a Live CD or patch a running system).

The produced TKLPatch can then be shared with the community, not just the ISO. This provides the ability for developers to collaborate on patch development, as well as reduces the work need to leverage the patch into creating an official TurnKey image and including it in the Turnkey library.

In the spirit of open source, developers are encouraged to share their patches with the community to help accelerate development. Once we add an appliance to the project we assume the burden of maintaining it at the appliance level with regular updates. This frees you to focus on more interesting tasks (e.g., improving quality of integration, software sub-components).

Nameserver in lucid core

I've noticed that when I boot up the VM with TKLCore Lucid I get no Nameserver from DHCP. Configuring from the console again fix the problem. Is this a known issue? 

Alon Swartz's picture

Yes, it's been reported

There have been a few reports (123) on DNS related issues, but we have yet to explore why this is happening in certain configurations, and in others not.

The issue will be fixed before the final release, but the more information you can provide, the better.

P.S - it would be best to move this discussion to the core-lucid-beta announcement.

Alon Swartz's picture

TKLPatch Live Training agenda

Just to give you an idea on what we'll be covering during the training session, here is the agenda:
  • Resources
  • What is TKLPatch
  • Development workflow
  • Anatomy of a TKLPatch patch
  • Live demo: prototyping Drupal5
  • Creating Drupal5 TKLPatch (walkthrough)
  • How TKLPatch works
  • Live demo: creating Drupal5 TKLPatch
  • TKLPatch development FAQ + answers
  • Open Q&A
Alon Swartz's picture

TKLPatch Live Training transcript

First, I'd like to thank all those who participated in the session - naudefj, Gonzalo, amoya, MarcBSchauber, slacker2, hoschi, ghoulman, Phani, Volkswagner and all the others who were sitting in the back.

We had a great time, though a lot longer than anticipated.

The transcript is available here for reference.

Jeremy's picture

Glad to hear it went well

I was keen to join in but the timing wasn't so greeat for me (3am Monday morning). Thanks for linking to the relevant irc log. I'll have a read a bit later when I get a chance.

thanks for the transcript

thanks for the training (I couldn't attend eventually) and the transcript !

Shameless plug

Novell has also launched a competition for develops to create Linux-based software appliances using Novell's SUSE Studio tool. You can post your appliances at the new SUSE Gallery, where users can vote on their favorites. Winners get a cool $10,000.

Here's where to start: http://www.novell.com/promo/suse/the-disters-contest.html

Liraz Siri's picture

Novell has met its equal. I raise you $100,000 and a pony!

Aha! So Novell has finally met its equal. Looks like you still have a little bit left from that $350 million you got from Microsoft to submarine Linux with patent FUD.

Brave of you to publicly acknowledge following in TurnKey Linux's footsteps. No need to thank us. You're welcome! Most soul-less corporations wouldn't have the cajones to position themselves as competing head-to-head with a tiny community-oriented open source project.

But I can't let you strut all those dollar bills around without a worthy response. It's well known that money is the prime motivator for everyone involved with open source. Nobody one-ups TurnKey Linux. You have Steve Ballmer on your side. We have Mark Shuttleworth. He's a filthy rich billionaire too. And much better looking than Steve Ballmer. And an astronaut!

So thanks to a generous donation from an anonymous, strikingly handsome billionaire, we're upping the ante to $100,000. Pony included thank you very much.

Now run along back to your corporate masters and ask them to raise the stakes or go home. Show us Novell still has what it takes to play with the big boys, because from where I'm looking that Microsoft-Novell 5 year plan to dominate Linux? Not going so well...

Yeah! My Pony Dreams Come True !!!

Awesome. I will get my pony and a check to cover the expense of keeping it "housed" and fed and available for free pony rides to neighboorhood kids. I think I better up my game a notch here.

Liraz Siri's picture

With that kind of loot...

You could take care of not just one, but two ponies. At the same time! According to my calculations this would make you twice as popular with the neighborhood kids. Not to mention the ladies, which I imagine would already be quite impressed with you for winning the TurnKey appliance development contest.

Just saying...

Liraz Siri's picture

New version of TKLPatch fixes Lucid + MySQL support

Many thanks to Adrian Moya who found a problem with tklpatch's support for MySQL under Lucid. If you're patching the Lucid TurnKey Core and you plan to use MySQL you're going to need to upgrade to the new version (very easy).

The new version also includes a few usability improvements so you may want to upgrade even if you don't use MySQL on Lucid.

Doubt

When i read the IRC log of TKLPatch Live training, i got one doubt.

I didn't find the command for creating the drupal5 DataBase. where actually it is specified ?

Alon Swartz's picture

During package installation

Just answered you on twitter, but for others who may come across this - The database is created during the drupal5 package installation (in the postinst script).

Liraz Siri's picture

Judging criteria

Basil sent me a private message asking for more information on judging criteria for the contest. I figured the response might be interesting to others as well.

As we wrote in the contest blog post judging will be based on total value added to the project and by that I mean how much value this allows the project to give back to the open source community and our audience.

We haven't thought in great detail about how to actually measure that value because it's still a bit premature and we also wanted to get the community to provide feedback on who they think provided the most value. Also, laying out precise "rules" for judging in advance invites people to hack the contest instead of hacking appliances and that wouldn't be very helpful...

But a few examples might help shed light on our thinking.

For example, say you work 2 weeks on one appliance but it's a highly specialized usage scenario (e.g., GIS database for topographical data) that will not help many users in the project's audience. That provides less value than the same investment in a more popular usage scenario (e.g., web filtering proxy or unified threat management).

As you note you can spend a given amount of time going deep (on one appliance), and going wide (on many appliances) or somewhere in between (a few high quality appliances).

That's a dilemma for us as well and it can be tricky to decide what to work on. From our POV which is better depends on which combination is likely to be more helpful to our users.

It's possible to help more users by going deep/high-quality on a few common/popular usage scenario than by going wide on a larger number of obscure applications. But it can also be the other way around. Circumstances matter.

There are a few things to consider:

  • TurnKey appliances are supposed to be turn-key. That is work out of the box while satisfying the user's usage scenario as closely as possible. An appliance that is buggy and hard to use creates a bad impression and may even damage the project's reputation and make it less likely users will try other appliances.

  • 80/20 rule - sometimes you can get 80% of the results with 20% of the effort.

  • Research the popularity of various projects using Google Trends:

    http://www.google.com/trends

    More popular applications are likely to be interesting to more people.

  • Since we're upgrading TKL to Lucid, TKLPatches that have been tested to work on Lucid are more helpful than TKLPatches that will not work on Lucid because it was only tested on Hardy. OTOH, a TKLPatch for Hardy is still much better than no TKLPatch at all.

Liraz Siri's picture

On the relative value of different appliance submissions

More thoughts from an out-of-band e-mail thread with Basil on the subject of how to add the most value to the project via TKLPatch submissions. Thank you Basil, these are good questions to ask.

I admit I wasn't familiar with all of the applications you've created appliances for (e.g., Efront, Tomatocart). When we get to the judging phase we'll research all submissions in depth but at least some of your submissions are for very popular applications (e.g., PrestaShop) that I expect there's going to be a lot of interest in.

Two tools we use to gauge an approximation of how much interest an application has are:

  1. Google Trends
  2. Google Keyword tool

For example here's a good trends chart that compares the interest in TomatoCart vs Prestashop:

http://www.google.com/trends?q=prestashop%2C+tomatocart&ctab=0&geo=all&date=all&sort=0

Here's one that compares eFront with Moodle:

http://www.google.com/trends?q=efront%2C+moodle&ctab=0&geo=all&date=all&sort=0

According to Google Keyword Tool:

  • 450,000 monthly searches containing the "prestashop" keyword
  • 1,300 searches containined "tomatocart" keyword.

Everyone should feel free to work on whatever interests them. But if you ask me what would add the most value to the project - applications which are likely to generate the most interest from users (e.g., downloads and installations). Submissions for better established more popular applications will probably do a better job at that then submissions for obscure projects with a small userbase.

It's true that TurnKey currently caters mostly to SMBs. That's a consideration, though for some purposes many enterprises use the same software applications as small to medium businesses. They just use them on a larger scale.

Here's a recent snapshot of the weekly download statistics so you can get an idea of what sort of applications people come to TurnKey for right now. Note how even our most popular appliance (e.g., Joomla) only accounts for less than 10% of total downloads. It's a long tail phenomena:


downloads: 4265

   ISO downloads: 1745 
   ZIP downloads: 2517

470  joomla-2009.10-2-hardy-x86           (zip:  187, iso:  283)
465  lamp-2009.10-2-hardy-x86             (zip:  284, iso:  181)
248  fileserver-2009.10-2-hardy-x86       (zip:  137, iso:  111)
216  drupal6-2009.10-2-hardy-x86          (zip:  157, iso:   59)
207  domain-controller-2009.10-2-hardy-x86 (zip:  114, iso:   93)
191  redmine-2009.10-2-hardy-x86          (zip:  128, iso:   63)
187  wordpress-2009.10-2-hardy-x86        (zip:  133, iso:   54)
169  core-2009.10-2-hardy-x86             (zip:   90, iso:   79)
167  zimbra-2009.10-2-hardy-x86           (zip:  100, iso:   67)
127  mediawiki-2009.10-2-hardy-x86        (zip:   83, iso:   44)
115  rails-2009.10-2-hardy-x86            (zip:   71, iso:   44)
112  otrs-2009.10-2-hardy-x86             (zip:   77, iso:   35)
91   mysql-2009.10-2-hardy-x86            (zip:   65, iso:   26)
89   moodle-2009.10-2-hardy-x86           (zip:   60, iso:   29)
83   torrentserver-2009.10-2-hardy-x86    (zip:   39, iso:   44)
79   openbravo-2009.10-2-hardy-x86        (zip:   51, iso:   28)
75   revision-control-2009.10-2-hardy-x86 (zip:   52, iso:   23)
72   postgresql-2009.10-2-hardy-x86       (zip:   42, iso:   30)
71   tomcat-apache-2009.10-2-hardy-x86    (zip:   45, iso:   26)
65   ejabberd-2009.10-2-hardy-x86         (zip:   38, iso:   27)
64   tracks-2009.10-2-hardy-x86           (zip:   41, iso:   23)
57   deki-2009.10-2-hardy-x86             (zip:   37, iso:   20)
56   tomcat-2009.10-2-hardy-x86           (zip:   42, iso:   14)
53   django-2009.10-2-hardy-x86           (zip:   33, iso:   20)
52   gallery-2009.10-2-hardy-x86          (zip:   35, iso:   17)
49   phpbb-2009.10-2-hardy-x86            (zip:   32, iso:   17)
49   dokuwiki-2009.10-2-hardy-x86         (zip:   37, iso:   12)
46   lapp-2009.10-2-hardy-x86             (zip:   32, iso:   14)
45   twiki-2009.10-2-hardy-x86            (zip:   34, iso:   11)
45   trac-2009.10-2-hardy-x86             (zip:   31, iso:   14)
41   bugzilla-2009.10-2-hardy-x86         (zip:   28, iso:   13)
39   bootstrap-2009.10-hardy-x86          (zip:    0, iso:   39)
38   projectpier-2009.10-2-hardy-x86      (zip:   22, iso:   16)
34   mantis-2009.10-2-hardy-x86           (zip:   23, iso:   11)
32   symfony-2009.10-2-hardy-x86          (zip:   23, iso:    9)
29   appengine-2009.10-2-hardy-x86        (zip:   21, iso:    8)
27   roundup-2009.10-2-hardy-x86          (zip:   16, iso:   11)
26   movabletype-2009.10-2-hardy-x86      (zip:   15, iso:   11)
26   moinmoin-2009.10-2-hardy-x86         (zip:   19, iso:    7)
17   ec2sdk-2009.10-2-hardy-x86           (zip:   14, iso:    3)

Some thoughts about this

My point of view is the following: IT IS important to view trends so to know about what are people looking for. That would surely help a project like TKL help more people and generate more value etc. But, on the other hand, new and wonderful open source projects that are build with fresh technologies and modern designs, and come to offer solutions to problems that old and popular and trendy apps have not addressed due to the difficulty of modifying their base code. So, a project like TKL could also help the opensource ecosystem presenting the final user with choices they were not aware of. And I mean "mature" or at least serious looking choices. You can test them first, see how's developing going on, lastest commits, see documentation, and finally offer appliances. That way TKL helps other opensource projects to grow, offering instant usage and the ability to test the software without having to follow complex (or not so complex) install procedures. They'll surelly be happy to post a link on their site to their app appliance, bringing their community to TKL. 

Basil posts have shown me good OSS that I didn't have in my radar. It's impossible to keep track of so much good OSS. That's why sometime when you are looking for a solution, I browse the TKL Gallery (and jumbox's I must admit) to see what appliances are there and follow the links to each project to finally download one that seems to suit my needs. 

Remember final users not always know what to look for. (That's why they end up installing windows anyway!). 

Does this makes sense to anyone? Just my two cents on this topic. 

Liraz Siri's picture

TurnKey adds value by aiding discovery of high quality OSS

I agree that a big part of the value provided by TurnKey is aiding the discovery and evaluation of quality open source software. There is some correlation between popularity and quality but as you point out it's not a perfect one, especially for new projects. If we can help these projects get discovered we're playing an important part in the open source ecosystem.

But all of this is only true so long as the noise vs signal ratio in TurnKey is low enough. For argument's sake imagine if we added TurnKey appliances for all open source projects. The low quality crap would drown out the high quality gems and users would probably end up going somewhere else for discovery.

So sometimes less really is more, but that's mostly a future consideration we're still far away from. There is still a lot of high quality open source software that isn't in the TurnKey library and that's something we would like to change with the help of the community.

Basil Kurian's picture

Can we have a similiar statistics about cloud deployments ??

The reason is that people may download appliances even if they are not interested in deploying it. Just to know how it looks and how it works.


Liraz Siri's picture

Good idea but we don't have that data yet

You raise a good point on testing vs deployment but the current batch of appliances doesn't provide us with that data. We could provide statistics from the Hub but it would be unwise to draw conclusions from that as the Hub is just a couple of months old while people have been deploying TurnKey in a much wider range of usage scenarios for a couple of years now.

I'm guessing there is at least a weak correlation between downloads and usage. If you don't download at all you don't use. The majority of downloads don't yet result in production deployments. I think most people still use the appliances for local development and testing but in the last year or so an increasing percentage of those have converted into "production" deployments that ask for security updates 7 days a week. That makes sense I guess. First people try, then when they're comfortable they "commit".

Deadline for submission

Liraz, cloud you please post a deadline for submission of TKLPatchs for the contest? By the initial post, it would be september 12? Just to have this date clearly defined. 

Liraz Siri's picture

Deadline is September 12th

But if people ask for a little more time to finish unpublished submissions or fix existing submissions we will consider extending by up to a week or two. So let's say by the end of September at the latest we should enter the judging phase. Does that sound reasonable?

Reasonable enough for me

Although I think two weeks will be a lot of time. Maybe one extra week it's enough. Just my 2 Cents...

Basil Kurian's picture

My 2 cents

The gap between first and second prize is too wide , it would be nice if the gap can be minimized laugh

 

2 methods are there ...... wink


Liraz Siri's picture

Dividing the pony would be gruesome

I see what you're saying but what would you do with half a pony?

Seriously though, I'm not sure whether you're kidding or not but just in case you didn't read it carefully our response to Novell was in jest. As a small open source project we can't really arm wrestle a massive corporation over money but we thought it would be funny to act the part. FWIW we're still giving away our real beer money though I imagine if you were actually expecting a suitcase full of money a thousand bucks first prize isn't quite as dramatic. Immortality in the TurnKey hall of fame, billions of grateful users and a cute pony should make up for it though. For the record if Shuttleworth ever does give us a suitcase full of money we'd use it to sponsor a dozen development contests not just one. Also we'd donate some of it back upstream to all the wonderful open source projects that make TurnKey possible.

Basil Kurian's picture

I thought it was real

cheekyStill no problem... I will continue my work .......


Liraz Siri's picture

Woops! Sorry about that...

If you missed the Novell comment and my response then it's easy not to realize what's going on. Just in case anyone else gets confused I added a link in the blog post from the prize money to the relevant comment...

Thanks for being such a great sport about it though...

Basil Kurian's picture

actually i'm a bit fooled too

I passed the information to many of my friends ... with a special mention to PRIZE MONEY :D

I just didn't get that joke earlier ..... (never went through the 'novell' section)

So now i need to correct that... "Because of less participation , they reduced the prize money "


Liraz Siri's picture

They'll just think you won and you're hiding it from them...

So when nobody's looking you pull in the suitcase from under your bed and throw fistfulls of dollar bills in the air and roll around in all that money laughing like a pig. I'm rich! I'm rich! :)

You don't want them to think that of you so you have my permission to tell them we spent it all binging on strippers and booze.

WHAT?

No suitcase with $100.000? How are we supposed to feed and maintain that pony? Oh well... I'll start a request for funds online right now! Do you have a cool picture of the pony so that I can post it?

Of Course !

Liraz Siri's picture

Low maintenance pony

Sorry about the suitcase (damn you Shuttleworth!) but don't worry about the pony. We've found one that is very low maintenance. Just think, if you play your cards right, it could be all yours...

Now that's what I call a motivator!

I can't imagine my wife's happiness if after so many extra hours hacking on tkl, I can get home with that lovely pony!!! No suitcase full of money can replace the value of this prize! 

Jeremy's picture

Yep, thats a pony you could be proud to own!

Show off to all your friends. And cheap to feed, easy to look after. It doesn't get any better than that!

comment subscription

Is there another method to subscribet to the comments, instead of asking? ^^

Liraz Siri's picture

Yes, if you have a user account

You can just click on the subscribe link at the bottom right of the first post in the thread...
Liraz Siri's picture

Adding $1000 to the contest pot

The contest is turning out to be a big success thanks to the amazing efforts of our two star TKLPatch developers (you know who you are!). We know the contest prize money is mainly a symbolic feel-good bonus for those involved the same way the occasional "beer money" donation is for us. It's usually not a lot of money but it still helps us to feel appreciated.

So as a symbolic token of appreciation for all the help our baby open source project is getting, Alon and I are donating an extra $1000 from our own pockets to increase the first and second prize:

  • First prize: $1000 => $1500
  • Second prize: $300 => $800

Thanks for helping us make the next release totally awesome. You guys rock!

Cool!

It may seem that $1000 extra compared to Shuttleworth's suitcase full of money is an insignifican't amount of money, but that colorful pony seems to eat a lot, and we surely will need the extra money to buy some combs, I was looking at these ones...

 

 

http://www.mlptp.net/forums/mlp-discussion/68112-any-these-pony-combs.html

 

So thank you very much!

Liraz Siri's picture

Time to start summarizing contest results

Autumn is here and with it ends TurnKey's wild and wet open source summer bonanza. Last night I finally finished reading the 300 or so forum messages I missed in the rush to release TKLBAM and I've started responding to the high priority items while simultaneously summarizing the contest results.

Note that this isn't like high-school where everybody has to drop their pens because the buzzer ran out. Refinements/fixes to the existing TKLPatches are still welcome and even brand new TKLPatches have a shot at making it into part 2 of the next release if they're good. Part 1 will come a little earlier and will include upgrades to the existing crop of appliances.

But as for the contest itself, it has to end some time and we're about a week overdue for that so it's time to start summarizing the results and ask the community for feedback (e.g., blog post, survey).

I'll be working on that today. If you haven't yet shared something you're working on with the community, now would be a good time to wrap things up or at least send me an e-mail so we can take everything into account.

What were the results of this contest?

I am especially interest if anything came up in regards to monitoring.

"Monitoring: Nagios, Cacti, ZenOSS, Hyperic, Zabbix, OpenNMS"

Jeremy's picture

The summary is here:

http://www.turnkeylinux.org/blog/contest-summary

But some of the patches that may be of interest to you are here:
Icinga
(Nagios fork)
Cacti

Salam

Druple Appliance and Ubercart

I am having issues using Ubercart on your system. Is it compatible? Will it work? THe issues seem to come when I try setting up my PayFlo Pro Account from Paypal.

Post new comment

The content of this field is kept private and will not be shown publicly. If you have a Gravatar account, used to display your avatar.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Allowed HTML tags: <a> <p> <span> <div> <h1> <h2> <h3> <h4> <h5> <h6> <img> <map> <area> <hr> <br> <br /> <ul> <ol> <li> <dl> <dt> <dd> <table> <tr> <td> <em> <b> <u> <i> <strong> <font> <del> <ins> <sub> <sup> <quote> <blockquote> <pre> <address> <code> <cite> <strike> <caption>

More information about formatting options

Leave this field empty. It's part of a security mechanism.
(Dear spammers: moderators are notified of all new posts. Spam is deleted immediately)