I have been reading about the Cloud Computing Interoperability Forum (CCIF) and its “Open Cloud Manifesto”. I went to their web site to read the manifesto but couldn’t figure out where it was hiding.

From what I saw on the site and the web discussions it seems that they would like to have a common API on top of the different cloud offerings so you can move from one cloud provider to another. On the surface this sounds like a good idea. It would reduce the risk of going with a particular cloud platform and make the different offerings available to a wider group of developers.

Maybe I have been around for too long but this sounds like a lot of similar efforts. There have been a lot of examples of vendors coming together on standards like TCP, HTML, HTTP, WS-*, REST, etc. but these have all come after the market has settled down a little bit. I can remember when everyone was talking about making sure that C compilers were compliant with the ANSI standard. Of course to make sure that there was a differentiator for their product every vendor added their own proprietary extensions. The same thing has happened with databases, languages, and even HTML as they were developed.

We already have interoperable standards like HTTP, WS-*, REST, and the work that has been done to make POX interoperable so I am not sure what they are proposing that would provide anything more than that unless they want to have a “least common denominator” approach to developing code in which case I can see this initiative slowing cloud development as the different vendors will have to agree what the least common denominator is and how each can implement it.

Until I can read the actual manifesto on their web site and not just reports of it I will withhold judgment on whether this is a good idea that hasn’t learned from history of if I am misinterpreting the news reports. In any case I think I will need to be convinced of this idea before I jump on the bandwagon.

A followup on this story. According to the cNet article at http://news.cnet.com/8301-13505_3-10206927-16.html, IBM was behind the drafting and promotion of the Open Cloud Manifesto. I guess that will doom the effort in politics and PR. I didn't bother to read the manifesto after I heard that it was a PR tool.

I have been looking at the Stonehenge incubator project on the Apache Software Foundation web site. I happened to look at their blog at http://blogs.apache.org/foundation/ and saw that today marks the 10 year anniversary of the founding of the Apache Software Foundation (ASF). In the past I have been critical of open source software. I still don’t get the business model but I do also the sharing of code and ideas as being much the same as what I do as a Microsoft Regional Director. I can respect the passion of the individual contributors and the desire of individuals to create high quality software. So I will say to everyone who has contributed to the success of ASF, congratulations.

If you are like me and couldn’t attend the MIX conference this year you can still get all the goodness. Most of the sessions are posted online and you can get them for free at http://live.visitmix.com/ where I can scroll down to the bottom of the page and see in the tag cloud that they have 129 sessions available.

I hate it when error messages have error messages. I got the error you can see in the screenshot below.

I closed the blank pop-up window over 50 times before I finally opened task manager and killed the instance of IE. After restarting IE I was able to get to the site and complete my work so it was a transient error.

The worst part of these errors is that it is difficult if not impossible to test all of the error handlers in all the ways that they will encounter problems in the real world. As a developer you might try opening the page, shutting down the web server, and then trying a postback. I would say that should be a normal test although to be honest I have usually done the test once and then did a copy/paste on the error handler so I didn’t test each one individually. The bigger problem come in when the error might occur half way through a response. If I get part of the javascript but not all of it there might be strange, spurious errors that would be very difficult if not impossible to test for.

I guess I don’t have a good answer on how to handle these problems but if someone does please add a comment and share how I we can all make our code better.

We will be having our February UCNUG meeting tomorrow at 6:00 at the NuSkin NOC at 1175 S 350 E Provo.

We will be discussing Microsoft’s cloud platform, Azure. Azure is an OS, set of services, and programming model that allows applications to scale as needed. We will be looking at the basics of the Azure platform and see how to write and publish applications to the cloud.

I am rolling the announcement of a couple of events into one blog post.

Salt Lake Cloud Computing User Group

The Azure user group will be meeting Wednesday February 18 from 6-8 PM at

New Horizons Training Center
2355 Technology Dr. (2355 South and 3420 West)
West Valley, UT  84119

Utah County .NET User Group (UCNUG)

Also on Wednesday, February 18 at 6:00 is the regularly scheduled UCNUG meeting. I was going to speak but just found out I have a conflict. I am trying to move the conflict or find another speaker. With this little notice I don’t know that I will be able to do either so stay tuned for an update next week.

Rocky Mountain Tech Trifecta

I will be speaking at the Rocky Mountain Tech Trifecta next Saturday, February 21. You can get more information and register at http://www.rmtechtrifecta.com.

MIX 09

The Mix09 conference is March 18-20 in Las Vegas. You can still get the $400 discount if you sign up today. Looking at the list of sessions I wish that I could go but right now it looks like I will be checking out the sessions online. You can get the full list of sessions at http://2009.visitmix.com/Agenda/Sessions.aspx.

In the latest edition of the Microsoft Security Newsletter there was a link to a series of articles on CIO magazine that discuss a series of claims from FireFox that it is the safest browser and the analysis of that by Jeff Jones who is a director of security strategy at Microsoft. I went back and read the first 2 articles and some of the surrounding discussions. While I don’t consider myself qualified to make a critique of the bug reporting methodologies or how to know which browser is more secure. Overall I think that security should not be a marketing feature but something that every company is trying their best to provide to us as consumers. It is hard and I know I have made stupid mistakes in the past so I am in no position to throw stones at anyone else.

One thing that caught my attention was the comments on the articles. There was definitely a lot of passion from supporters of both browsers. Some of it might be attributable to “fan boys” but I think it goes deeper. With so much of our lives moving online the browser becomes the platform we use to interact with the world. As long as 10 years ago I can remember having discussions about whether Microsoft would be selling an operating system in 2 years. Certainly if everything moves into a browser the OS will become irrelevant to a large extent. I don’t see that happening any time soon since there are still many user experience reasons (like shortcut keys and the ability to keep client state) where “fat client” applications are easier to use. These problems are not easy to solve so I don’t expect to see a fix soon. However, I can certainly see where people are passionate about their computing platform because it influences so much of what they can and can’t do easily and to a large extent their thinking of how to write programs and interact with the online world.

I have been looking into Windows Azure lately getting ready for the technology and for the talks I am planning on giving in the near future. When I was telling one of my friends (who loves everything Apple and loves to bash Microsoft) about Azure he was laughing about how the “cloud” sounds very abstract and hard to understand. In the end he was equating the “cloud” to vaporware. I hope I set him straight about how the cloud is more like the Internet where he has data on content delivery networks.

Today I read this article about someone who’s credentials were stolen through a phishing scam and his information and community on Orkut were changed by the phishers. He tried to contact Google about the issue but couldn’t find anyone to help him fix the problem. That has had me thinking about what happens when data or applications have problems in the Azure.

Microsoft has been very clear that the current implementation is pre-release and before it becomes a “real product” they will have a SLA in place. I don’t know what that SLA will entail but I suspect it might be something like Hotmail. Right now I am paying a small yearly fee for Hotmail so I can send and receive larger files and I don’t have to worry about my account being disabled due to inactivity. At one time I had a real need for these services and the price is so low that I keep paying it. What really got me hooked on paying was that I could use Outlook to retrieve my e-mail. This is so much faster and more convenient than the web interface since I have Outlook open anyway. I also found that it got me some enhanced support options (which is really more relevant to the purpose of this post). When I was having problems there was a dedicated support alias that I could use and I got a reasonable response from a live person. I haven’t tried the support on a free account but I would suspect that it would be more automated and point me to the FAQ. I can live with this since I really shouldn’t expect Microsoft to pay for bandwidth, storage, and tech support for a free account. I see trading good tech support for free e-mail. It is a risk/reward equation where I either feel I can fix it myself, ask the community for support, or just drop the e-mail account and start a new one.

If the Azure SLA follows the Hotmail model (and I hope it does) there will be different levels of support depending on how much you are willing to pay. As your applications and data in the cloud become more important you will be willing to pay more and expect more from Microsoft like better response times, a knowledgeable human being to help answer questions, and higher levels of redundancy and safety on your applications to avoid problems in the first place. I don’t know if there will be a free service but I hope there is for things like small applications that hardly use any resources. At that level I don’t expect Microsoft to do anything if I tell them I lost data from blob or table storage. If I am paying a modest fee I would expect them to look at the problem and help me recover to some state in time. If I am paying a lot then I would expect “up to the minute” data recovery. I could live with a model where I pay for the protection I want.