Chrome OS, what is the big deal?

Goggle announced the Chrome OS a couple weeks ago. I have looked at it, watched the videos and there are some short comings that I see. 1) When you are offline, you might in trouble. Since your documents are online and not stored locally that is a problem. 2) Google says you don't have to worry about "Nagging Updates". But here is the problem, if a web app is updated and there is a bug that trashes your data, you are screwed. Either your data is trashed or you find out before hand and you can't use the app until it is fixed. 3) The problem of trust. Do you trust your data to Google or whoever's web app you will be using? It is not just trusting them with your personal data. But trusting that somebody will not hack the provider and trust that they will not lose your data. A reminder of the problem with trusting a single source is the whole Sidekick/Microsoft fiasco. Will the provider backup your data properly? What happens if the company goes under? Can your data transferred to another provider? The funny thing is that there will be some information that is stored locally. The question is how much? HTML 5 allows for local databases. What will be stored in those? Will Chrome have to store the web apps and data locally and then sync the changes so that you can use the Net Book w/ Chrome OS when the Internet is not accessible?

Now some of the things are quite interesting. Like the booting process and the patching process. It reminds me of Solaris, in that you have 2 copies of the operating system one that is patching the other and then if the patch is successful on the re-boot it will switch which copy gets booted from. And if the patch fails, there is no damage. I also, like the verification process, this can mean that any malware will have a tougher time. I hope that the OS will recognize the difference between a root partition and a user data partition. So that nothing in the user data partition can be executed. This should make it harder for malware.

When it comes to the UI, it looks nice though the one thing that needs to be improved before it is released is the way external storage devices are shown. The question is: Will we be done with using tabs by the time it is released? I am thinking of using a fan-type UI that shows circle possibly that then could be rotated to select the window/panel that you want to see.

Now, do I think a Chrome OS NetBook would be useful? Yes, I do. An example currently, is the family "living room" laptop. It has Open Solaris installed on it and the main reason for it is to allow my kids to browse the Internet, watch videos and play games. All of these are on the web. So a Chrome NetBook would be a perfect match for this situation. So I think the Chrome OS has a future and it will see how it is marketed to the general public and not just us geeks.

Oracle buys Sun. What happens next?

Oracle is overall pretty predictable. Buy the company that has the best product for in that segment of the software industry and then drop what products that are duplicate but inferior, including their own product. (e.g. dropping OC4J and replace it with BEA's Weblogic) But the purchase of Sun has me scratching my head.

Yes, I am happy that Sun was acquired by Oracle, instead of IBM. I think IBM would have been a mediocre partner. They would not do what Sun needs. Sun needs to review what it has and drop what is just a distraction. It will be interesting. Oracle normally just integrates the company into their organization. But those companies normally don't have the variety of products that Sun has. And there is an additional thing that Oracle has not had to deal with in the past, a hardware-based division.

Every company (that I know of at least) has been a software-based business. But Sun sells hardware in addition to software. But also, the current revenue model that Sun is using for its software is mainly a service-based model. (i.e. free software, but you need to pay for support). Obviously, it has not worked out as well as they would like. But that is not the only problem for Sun. They had some missteps with their acquisitions that hurt them too.

So what happens to everything that Sun owns? I don't know for sure. Heck, I don't think Oracle knows. But lets look at some of the offerings of both, where there would be duplication.

NetBeans vs. JDeveloper, I think NetBeans is the better product and supports not only Java, but PHP, Ruby and Python and Oracle is definitely interested in all four of the languages.

Solaris vs. Unbreakable Linux, this is a tough one. I think Oracle would love to focus on Solaris, because it would finally have an OS that is theirs. One that they can control and make available as an all in one solution, from top to bottom. And one of the advantages to Solaris vs. Linux is that Solaris supports Containers (a light weight way to have a protected environment without a full VM) and more importantly ZFS (which is the future of file systems). Also, support for x86 has greatly improved in the past years. Given that I think Oracle will drop Unbreakable Linux and put it OS effort into Solaris. But Oracle will fully support Linux, but I think Oracle will see the benefit or concentrating on Solaris.

MySQL vs. Oracle's Database, this is an interesting one. And to be honest, I can't imagine Oracle killing either one. MySQL and Oracle's Database have two different markets. The only thing I see happening is that Oracle might provide a better migration from MySQL to Oracle. MySQL is a great starter database. But there is a point when people realize they need to upgrade to something more powerful with better transaction support.

Java vs. JRockIt, this one is not really about picking one over the other. But the question is more of will Oracle take some of the changes that BEA did to the JVM and merge them into the official Java implementation. And of course, will they continue with the JSR process. Also, it will be interesting to see if Oracle will push for JSR 277 to be implemented as a pure OSGi implementation.

Glassfish vs. Weblogic, this is one that I think Oracle will choose Weblogic. Oracle will probably help to form a group that will take control of Glassfish and leave it to the Open Source community. I have a hard time believing that Oracle will change J2EE containers again. And then there is maturity of Weblogic vs. GlassFish. Glassfish track record is not proven compared to Weblogic. Hey, for all I know Oracle might see how well Glassfish is doing and drop Weblogic. But I have a hard time believing that. Especially, with the direction change that Oracle has done with moving BPEL from OC4J to Weblogic.

I am sure that there are other products. But they don't come to mind.

CompUSA is back?

This will be quite interesting. The question is: Will people trust the "new" CompUSA?

CompUSA closed its doors 2 years ago. People ran to the stores for the "Going Out Of Business" sale. They got good deals on those items with no expectation of being able to return them. Now, they are saying "We're Back!!!". What the heck?

If you are going out of business, do it and get it over with. Liquidate start a new business. But don't pretend that nothing happened.

Did the iPhone get it right?

I have been reading Chapter One of "Palm webOS" and it looks quite interesting. They have definitely done things that I like, like requiring Palm apps to be signed. I assume that the certificate will need to come from Palm, like Apple did. That would be the way to go. It will be interesting to see how far the "web" metaphor will go. To be honest, I have never designed an interactive web page, except some minor use of div to hide and show parts of a web page. I have seen Google Calendar and things like that and go "Wow!! How did they do that?". Somethings like making the "native" APIs available via JavaScript were expected after I heard that the "webOS" would be using just web technologies. Of course, it makes me wonder if some of these "native" could be made available on other smart phones or desktops so that the webOS apps could be available there too. As I said interesting.

Though, the more I look at the webOS, the more I think that Apple got some things right with the iPhone. The first thing that hit me was background applications. Because webOS (and some other smart phone OS's like Android) let applications run in the background this will add complexity to managing the device. Think about it. On a small device like a phone do you really work in multiple applications at the same. The answer is no. You usually open one application work in do what you need to do and you then switch to another application. You are not really work in the two applications at the same time. Now, one of the things that Apple recommends developers do for their iPhone apps is that when the home button is pressed and application is exited, the application saves the state of what the user was doing. Why? So it when the user goes back into the application it is as if they never left. What does this do? This simulates the running of multiple applications at the same time without consuming more resources.

So what is the problem with allow those apps to just stay running in memory, you might ask? The problem is that they consume limited resources, drain the battery faster and most importantly cause the user to have to think. You might say "Thinking?!? Thinking is a good thing." Overall, yes thinking is a good thing. But the problem with having multiple applications running at the same time is that you need to start making decisions about what applications you leave running and the applications you actually want to completed exit out of. There was a study done a awhile ago that should that having a choice was good and made users happy. But if there were too many choices, then the users got confused and very unhappy. This is the power of simplicity and what the iPhone got right with not allowing applications to run in the background.

The second thing that I think the iPhone got right is that there needs to be a kill switch for applications. Now, it is possible that it is in the plans and it is just not mentioned, just like Apple did not mention their kill switch until it was discovered by somebody watching the traffic of an iPhone. I think this is more critical to the webOS than the iPhone. Because (outside of jail-breaking your phone) background applications are not allowed. With the webOS they are allowed so it would be more important to have this to protect again rogue apps.

Overall I think what Palm has done with webOS is really cool. I can't wait to see how it turns out.

Stupid Idea from the County

I heard a rumor the other day. Something, that just made me shake my head. Ok, currently one of the agencies of my county government currently has multiple private providers to help service all of the people that they are trying to help. Now here is the stupid part, they are thinking about (I think they might being pushed by the State government) to select single private provider and just contract with them. What is the problem with that? Simple, if you use a single provider it does not give you any flexibility. You can't switch to a different agency if one if not up to snuff. What happens if the agency slacks off and does not perform their duties properly. You are stuck in that contract for the rest of the year. Yes, you can add performance requirements. But what happens to the rest of the private providers during this time? They are gone, closed down, kaput. What are you going to do then? Nothing because you no longer have a choice. You have a system of great flexibility by having multiple providers. You give many people work in addition you carry a very big stick on a month by month basis. If Provider A is not doing well, then give the new cases to Provider B or C, this will hurt Provider A and gives them an option, improve or die. The best system is a system where there is constant competition.

Savior for newpapers/magazines

I was just listening to Leo Laport of Leoville.com he suggested the New York Times give away the Kindle to save money. This got me thinking and given the cost of printing and distribution of newspapers and magazines, I think the Kindle could be their savior. The Kindle could save the print industry.

What is the greatest concern about having newspaper available either online or downloadable? Theft. People being able to read the newspapers without having to pay for them. What is the solution? The Kindle.

What is the one thing about the Kindle that makes it such a great solution? It is a closed system. People can subscribe to the newspaper or magazine and have it delivered to that and only that Kindle. This makes a perfect solution. All the newspaper industry needs to do is subsidize the Kindle and sell it for $49.95 with their subscription. Keep the cost of the subscription the same. This will more than pay for itself.

There are at least 2 problems, I see:

1) The limited number of Kindles would be exceeded very quickly. This should be solvable by having the newspapers make deals with Amazon which allow them to ramp up production with little to no risk.

2) Not everyone wants an electronic edition. They like the feeling of paper and that is what they are used to having. The question is really, how large is that contingent and is it small enough that you can afford to let them go for the greater good of the newspaper?

But overall, I think this is way newspapers (and possibly magazines) will need to distribute their content in the future.

Switch to Digital TV Delayed!!!

Yesterday, Congress and the President did something very dumb (Big Surprise) they delayed the switch to Digital only broadcasts from February 18th, 2009 to June 12, 2009. In a time of economic crisis, the government just pulled the biggest bone head play. They just cost all of the broadcasting companies, the small stations that barely have enough money to pay for expenses include some NPR stations to the largest broadcaster, millions of dollars. Now, what will the result be of the delay? Well, depending upon the broadcaster they will either need to close up shop because they can not afford to stay open or cut staff and possibly other programming to make for the additional costs. Is this what we need at this time of economic crisis? No.

If Obama blames, the jobs lost by this move on Bush or the Republicans then he will be a liar and just another finger pointer that is just trying to deflect blame. Obama has been pushing this since the day he came into office. Yes, AT&T and Verizon will not be effected since they decided to extend the leasing of the frequencies by 4 months, but they probably weren't going to use it right away anyway.

The thing is no matter how long they delay they will be millions of Americans that will not be ready and millions that will be losing TV broadcasts because they are on the fringe of TV signals. And will the delay really change the number of people that will be affected that much? Probably not.

Total Ripoff

If the above link is true, then there should be an investigation by the Federal Government on this truly outrageous tactic. That is taking the current prices, jacking them up and then say they are discounted.

Heck, why even do that. Just put 20% stickers on everything and not change the price at all and say that the price is already the discounted price. Gotta love it when the vultures are buzzing around. Unfortunately, they are buzzing around the wrong thing.

Why did Circuit City fail?

I think there were two reasons why Circuit City failed.

1) Their idea for pickup at the store after ordering online was a great idea. I for one was one of the people to use the service and I loved it. I could order from work and on my way home just stop by and pick up my order. Great idea. Problem is that they dropped the ball. This was a strength, but they did not adjust to improve the service. So BestBuy took that business. BestBuy has done a pretty good job. Yes, they could improve and I have sent many ideas to them via their satisfaction surveys.

2) Circuit City did not focus in the right products. Circuit City seemed to dabble with computers peripherals but the selection was short. They had a good selection of the computers themselves but either they should have removed the computers peripherals and DVDs and focused like Tweeter. (i.e. the big ticket items and look to keep the prices down for those)

Circuit City basically failed because Best Buy did it better. They were larger and did what Circuit City did but better.

Mercurial is awesome

I have been using CVS for my version control needs for my business for many years. But I wanted to move forward to something that would give me more flexibility. I originally looked at Git, but it seemed to complicated. You need to run a garbage collector ever so often to clean out the crud. I don't want to have to remember to clean up. Also, I looked at how many commands there were and figured that wasn't for me. And the straw that broke the camels back, was that seemed to be a hobbled together set of scripts and binary code.

I looked at Mercurial and liked it. It was all written in Python except some extensions written in C. I liked the approach. I looked at the number of commands, only 16. Nice simple, yet powerful as Git. I liked it. I have started to use it and it is cool.

Dirkjan Ochtman sent out an Email asking for help with slave build machines. I thought about it and decided to volunteer my iMac. Well, I like NetBSD and FreeBSD, too. My favorite being NetBSD for its simplicity/bare-bones approach. So I thought about grabbing some of the my boxes I had laying around and set them up. Well, I am lazy and I love to use VMWare's Fusion to run different OSs. So I installed both NetBSD and FreeBSD in different VMs and I checked the memory usage. It didn't seem to high, so I decided I could volunteer both of those too.

So if you go to BuildBot: Mercurial you will find my three "machines". It is cool to be able to help out. And what is even cooler is that I was able to do it on my "server" Mac.