JoshSweeney

Open Source: Contribution Based Upgrades

After being heavily involved in the SugarCRM community and the communities of other open source software over the years, there is one thing that has started to really bother me. In the BitTorrent world we call them leechers. These are the people who take and never seem to give back to the cause.

The problem that I have with this is that people are continuously leveraging others hard work while many times refusing to pay for anything attached to it. After download the platform they get the free modules, then they want consulting and haggle about the rates trying to make a point that the software was free stating “why would I spend so much on consulting when the platform came at such a little cost”. Lets face it, peoples time deserves compensation.

In building WapSnap and GetSocial Twitter Pro I thought long and hard on how to give to the community without rewarding the leechers. This leads to the idea presented in the title, “Contribution Based Upgrades.”

The concept is that the software is open source but to get the upgrade packages you need to contribute. This concept isn’t without it’s hurdles but we will get to that next.

Here is how it works. First off the software creator will require the user to sign up to download ( Even though I hate this concept ). During that process they will be told that they will not receive upgrades without contributions. The software provider will maintain a portal with a forum, documents area, and any other content sections needed to support the software which can take contributions from users. When a user posts any content, they will get points from the system or other community members with minimum and maximum points allowed based on the type of contribution.

The catch is that they must maintain a certain level of contribution averaged over time to have access to the upgrade packages. The level doesn’t have to be high but the fact that they have to contribute in some way is a step in the right direction. Maybe they could even get extra points for external blogs posts, who knows.

Now for the problems.

1. First is that current licensing of open source software will allow one person to download and distribute the upgrade packages without any recourse. Not that you want to go after users but you do need a “leg to stand on” if needed. To fix this you will have to draft your own license agreement.

2. Since the system is points based you will have to make sure that people aren’t gaming the system. This will probably be a constant battle for an active project but as long as your cover the majority of issues it shouldn’t be a huge problem.

3. People posting garbage. If people continuously post garbage to maintain there status the community will usually report it. If not, the moderator can troll for garbage and take it to the curb.

I’m am sure that there are dozens of other issues that could arise but at least the list is started.

In theory, this would greatly enhance the project quality and allow users to give time back to those who so graciously gave to them. I am interested in your thoughts on contribution based upgrades so send them to me or make a comment.

Tags: bittorrent, contributions, open source

An Introduction to WapSnap

I have been actively Tweeting about WapSnap and speaking to various individuals about what it does. Today I determined that it was time for a proper introduction.

WapSnap meet world, world meet WapSnap.

Many of the intricacies about  WapSnap’s product positioning are still in limbo. For now all that I can say is that it is a web based snaphsot and comparison utility that will compare both files, and databases to detect changes. Most peoples first response has been,  “doesn’t Git and SVN do that already?” My answer is, yes but only for the files.

The file comparison will be similar to a WinMerge and other comparison utilities like those built into Git. Where I would really like to differentiate is in the database comparison and manipulation area.   WapSnap will be able to compare databases all the way down to the field level in order to detect changes in applications. This leads to why I started building WapSnap.

Although others and myself attempt to use best practices when developing applications, sometimes developers end up working on the production box instead of doing an update from their favorite repository. This increases the risk of error and possibility of breaking production applications. The first response to this is to restore backups and risk losing data entered since the last backup. The alternative to backups is to track down the issue and try to fix it as quickly as possible. If you try to fix it the question will become, what all was changed? This is where WapSnap comes in. If you snapshot both the files and database before the changes are implemented then you narrow the search area to check for errors.

This becomes even more helpful for those using open source applications like Joomla and SugarCRM. Through many discussions I have found that users of open source applications commonly work on production instances because it is easier, cheaper, faster, and the applications take plugins or modules that most think work out of the box and won’t break their system. Yet another example of WapSnap’s handiness is when someone installs a module, it breaks their app, then they uninstall and the app is still broken. WapSnap will easily let you detect the crumbs left behind in the file system and database so that they can be removed in less time that of what is needed to do a large restoration from backup.

Another huge benefit is that you learn more about the application. Take SugarCRM for instance, does anyone really know everything that happens behind the the scene we you use the Module Builder to build and deploy a custom module? Most people on the SugarCRM forums don’t and neither do I, most just speculate and build on past knowledge to guess what it is doing. The same goes for deploying 3rd party modules in any application.

Last but not least, where in the world does WapSnap comes from. I needed a name for a project that I was working on that took snapshots of web applications so I decided on W-Web, AP-Application, SNAP – Snapshot. Nothing special but it is staying until I come up with something better or decide to stick with it.

If you know of other tools that are helpful that have similar functionality please comment and let me know.

Tags: dev tools, open source, snapshot, WapSnap

BarCamp Atlanta 2 – The Final Hours

The final hours of BarCamp was a testament to how much everyone wanted share knowledge and learn from each other. Everyone was tired, some hung over, and many just waiting to go home and pass out for a day or two. Either way, here is an overview of the last sessions that I attended.

After lunch I attended Russell Journey and Loren Normans overview of Amazon Web Services. We received a quick overview of Amazon EC2 and then talked about when to make the switch from your standard hosting environment. The consensus seemed to be that you never actually make the switch. EC2 and other services are really built to assist in making your application fully scale when traffic starts to flood in. One example use is to have you website configured to spin up new EC2 instances when a certain load is met. With the correct structure EC2 is a cheap alternative to building out your own infrastructure and can easily handle getting Slashdotted or making the Digg front page. The second part of the speech was a flurry of URL’s to help developers more easily handle instances and cloud services. Here is a list in case you are researching cloud services and management. It also include some of the audiences preferred web hosts

• bluehost.com
• slicehost.com
• ylastic.com- I believe someone said these guys were local.
• rightscale.com
• scalr.net

Next I jumped into a security talk from Shauvik. He was giving an overview of the latest web security, projects, and updates. He went over various FF extensions used for auditing and some tools from OWASP.

Tools:

• FireCat
• Live HTTP Headers
• Wikto
• Nikto
• W3AF

Following the security update I sat in on the Mini Startup Gauntlet session paneled by Lance, Sanjay, and Don. For those who are not familiar with Startup Gauntlet it is a reoccurring event in which you present you startup pitch and the panelist tell you how to make it better. During the mini version there were presentations by Russell Journey, Mike Mealling, and Tejus.

Tejus completed his pitch of SCMPLE and I had to rush out to give one of my presentations which was Attack Detection and Remediation using RIA’s. I will have full posts up on my presentations shortly.

Now it was time for the last session of the day. Actually, it was the second to last but it went so well that most people turned it from a 30 minute session to an hour. Doug hosting the open discussion about tagged Open Source: The who, what, when, where, and why. Since most people in the room could fill in those answers pretty rapidly we attempted to think on a deeper level about open source. By doing this we really expanded the conversation. Here are some example answers that were written on the large white papers surrounding the room.

What: Other than software what are other uses, copyright and media,

• Who: attributes of who? Customers are architects and devs, freelancers, cheapasses, savvy cio’s, “do it yourselfers”
• When: when to use the Open source structure/arch?
• Where: everywhere
• Why: cost savings, community / collab, learning, better and easier to maintain, ready made consulting markets,
• How: more freelance help, google code
• How much: is open source free, jboss was purchased for 360,000,000 , mysql was bought for 1.2B, redhat 30% yr over yr growth for 4 years running,

As you can see, many of these were more questions just to expand the topic. We spent much time analyzing companies that were open source/free, open source enterprise, commercial open source, and other forms of hybrids. Overall my biggest take away was that the word open source has really gotten skewed over time. In the past and even today many people treated open source as if it was synonymous with free. For a period of time that was somewhat accurate. However, some of the largest open source companies today don’t have as much to do with free software as when they started. Overall this was a very compelling end to BarCamp.

Lastly I have to thank everyone who made it happen.

If it weren’t for Mike Mealling and Lance Weatherby this would have never happened. Thank you for a great event.

I must also thank the BarCamp sponsors that make these possible.

• Angus McRae
• Appcelerator
• Microsoft
• Ryma
• ATDC
• GeorgiaTech Venture Lab
• TechLinks

BarCamp Atlanta 2 Links:

• Pictures from Lance
• Why Erlang Slides
• BeagleBoard
• Web Security
• Generating Buzz ( Without being a scumbag spammer )
• Push Protesting
• How to build your own Guitar Hero/Rockband clone in a week
• Hacking Ubiquity
• All of the Tweets

Tags: amazon, atlanta, barcamp, cloud, open source, security