Kaamel's blog has moved to http://kaamel.kermaani.com

Cloud Independence Ahead ...

noreply@blogger.com (Kaamel Kermaani) — Fri, 04 Sep 2009 20:30:00 +0000

Cloud Computing is getting a fair share of technology news these days. In a nutshell it is a way of providing and/or using remote computing resources through internet. The business model is evolving but one way or the other the end users pay for the usage of the resources instead of purchasing additional equipment. This is of interest to the hardware vendors as they would be able to sell their products as a service, which means a constant flow of money, albeit in smaller chunks than selling hardware outright. Their vision is that they can then control the deployment of new hardware and not rely on the end users to make hardware purchasing decisions directly. To make this work, and to make it work seamlessly, layers of software is needed on the end-user side and on the remote resources.

Like many other technology advances, in the first phase the cloud computing vendors needed to provide the pieces since there was not much else in place. As a result the current vendors each have their own solution that to a large extent is not compatible with others, making migration or multi-vendor configurations difficult or impossible.

Deltacloud, a new open source project within Red Hat, announced recently, is trying to solve that problem by providing a "cloud broker" that would convert the communications between the consumer end and the cloud end of one vendor to another, making the interoperability possible (see the diagram). There goal is to enable an ecosystem of developers, tools, scripts, and applications which can interoperate across the public and private clouds.

While Deltacloud is a young project and more than likely will face other competing products, the problem that it is attaching, the interoperability, is a major requirement for users, developers, and IT departments if the cloud computing is to grow rapidly. Eventually, there will be a very small number of (or ideally just one) standardized means of connecting to the cloud services (also referred to as API or Application Programming Interface), but in the meanwhile, this translation layer is the next step before the standardization of APIs.

To a large extent, after it is all said and done, to the end users, this will be transparent and invisible. At the end the average user is interested in the end-services that can increase their efficiency and productivity. The lower total cost of ownership has been the primary selling point of cloud computing. But the low cost has been the norm and not a new trend. In my opinion cloud computing needs a killer application to become a difference maker, otherwise it will remain as yet another step in the ever evolving technology quest.

And that is how I see it ...

Time for My Monthly Vision Thing - Flash as a Computer: FaaC

noreply@blogger.com (Kaamel Kermaani) — Sun, 30 Aug 2009 21:08:00 +0000

I am always thinking. I don't know if it is a good thing or not, but it is something I cannot stop, and frankly why should I? After I am gone it will stop forever so I might as well let it go while it wants to. So I have decided to write about some of my crazy ideas once a month. Here is my first ever monthly vision thing.

I think the time has come for what I refer to as "Flash as a Computer" or FaaC. I must admit it is not entirely a new concept, but my twist to it might be somewhat different, or at least it is not based on what I have read or seen anywhere else.

We have all seen the flash memory cards, right? Those little memory modules that we use in cameras and increasingly in camcorders. Their capacity is getting large enough (I have seen 64GB in stores) that they are now comparable in size to the disk drives in some laptops. And that comparison triggers the thought that maybe the entire computer environment can be saved in one. In other words one way or the other one might be able to hold most if not all of the setups and unique data that is ordinarily on disk in a laptop in a large capacity flash memory card. Then why can't we have a new generations of computers and laptops that can read the settings and data off these flash memory cards and make the computer act as if it were our own computer. If instead of carrying my own laptop I have a flash card that I can take with me and stick into any available computer in a hotel, in the office, airport, or wherever, and if that computer could act just like my own computer, it would be very convenient, don't you think? Probably one would have one FaaC for home computer and one for the office, and maybe for practical reasons, these FaaCs could be made the size of credit cards. I found the idea intriguing.

But is this practical? One potential approach would be to replace the boot disk of one's computer with a portable flash memory card. Since there are many different chipsets out there and such disk would not necessarily work everywhere, however. So what if we can create a virtual system on the flash and that way it would work on any computer that can load that virtual system. And if the ability to load a virtual machine is built into the hardware (it is technologically doable today) then we have a solution. And there can be a few enhancements too. For example there could be a storage hierarchy. Sensitive or archived data can be stored on a secure local disks. These storage devices would be accessible only locally, some at home, some at work, etc. Once disconnected they would not be available which is desirable. At the same time it would reduce the required store space on the flash memory card. Some other data, the ones that are not very sensitive and also not necessarily needed all the time can be stored remotely via internet (also known as "in the cloud"). These are internet based storage spaces like "SkyDrive" offered by Microsoft. An example of files that could be stored in the cloud is the old multimedia files that we no longer listen to or watch on a regular basis, but we don't want to throw them away either. They would be available only when online, and that could be a reasonable tradeoff to further reduce the need for larger flash memory cards. The files that are used frequent enough can be stored on FaaC and they would remain available all the time. A clever piece of software could monitor the file usage and make sure the most recently used one (the last 30 days?) are moved to FaaC (actually copied and kept synchronized with the original files). In addition, before getting on the road one could manually "take" additional files and data that might be needed while offline, on FaaC.

Software licensing would be the next issue. Currently when there are two computers at home and five users, there are only two licenses (of Microsoft office, for example), one on each computer. Would they need 5 copies with FaaC? The cost could be a problem. There needs to be some way of associating all those FaaC with the available licenses (two) and be able to have only two of them authorized at any given time. This is possible, using some sort of internet based authorization, but what happens when the machines are offline? May be they can check out a license before going offline? No matter what, the software vendors need to embrace the concept and come up with workable options and solutions.

So what does it take for this to become real? The hardware vendors need to agree on and implement a hardware based virtualization, software application vendors need to sort out the licensing options, the operating system vendors (Windows, OS X, Linux, as a minimum) need to make the FaaC concept and the overall storage hierarchy work seamlessly and transparent to the users, and the businesses, hotels, airport lounges, etc need to all buy into this concept and deploy the base hardware and software. Somebody needs to pull the trigger and make others follow their lead. Somebody, please? To start with, maybe the Linux community could create a stripped down kernel that in essence would turn an existing hardware into a system that does nothing other waiting for a flash memory to be inserted into at which time it would look for and run any virtual machine it could find on it. This is not fancy, but is enough to allow people to take an image of their systems with them and run on other hardware. If the concept proves itself, I am sure money will follow.

And that is how I see it ...

Will cost cuttings result in more jobs to be outsourced?

noreply@blogger.com (Kaamel Kermaani) — Fri, 28 Aug 2009 19:43:00 +0000

This is a repost of an answer I recently posted on Linked.

I think in the short run no, in the long run yes.

Outsourcing, especially to other countries, started more than a decade ago (in its current definition). It was overdone and rushed, before creating the right processes and procedures to measure, monitor, manage, and adjust. There were some successes and lots of failures too. The failures slowed down the practice and in some cases even reversed the trend. As methodologies became more and more established and better implemented, and as the vendors learned to meet the needs of their clients better, the outsourcing started becoming the norm rather than the exception. and I believe this trend will continue.

From a bigger picture perspective, to remain competitive companies need to do things horizontally (focusing on their particular expertise and added-value) and outsource functions that are required to run a business but are not particular to their company. This is not necessarily bad for the workforce, although it does shift the jobs around. Also this does not necessarily mean jobs will move outside the countries, although in some cases they do, but for some other jobs there should be an increase via insourcing (if that is a word). This process, over time, should level the playing field.

To Be or Not to Be (online)

noreply@blogger.com (Kaamel Kermaani) — Wed, 26 Aug 2009 21:59:00 +0000

Many small businesses believe they don't need to be online. They feel it is complex and/or costly. Most of those who are online, don't realize enough real benefits. The fact is, more than ever businesess of any size need online presence. For example a one-employee company (maybe a consultant) might want to have their resume, past clients, expertise, and contact information on a website that is bookmarked by clients or prospects, in addition to being searchable when others are looking for someone with their background. A small local store might want to offer coupons, specials or other information that would connect them to the community. Furthermore, online presence is not expensive, time-consuming, or too difficult for a small business, and with a little planning and foresight it can effectively increase market penetration which often results in higher revenue.

Here are a few tips to get online, successfully.

1. Start with a website

There are quick and easy ways to get online and have a presence. Depending on the nature of your business, you could need a more interaction with your partners, clients, and prospective customers through your website, in that case you might want to consider getting professional help to create an original design that sets your company and business apart from the competition.

2. Get your own email address

As a business, your email address should not be from yahoo.com or comcast.net. Your emails should be originate from your own company domain address, for example John.Smith@HisOwnCompany.com. This goes a long way in establishing credibility with everyone you deal with.

3. Host your website based on your needs

You will need to host your website somewhere. There are many hosting companies but there are tradeoffs just like anything else in business. The primary factors include storage capacity (how much data you will have on the host), access speed (how fast the pages will load in the browsers), reliability (what percentage of the time the host is operational), flexibility, and cost. If your website is frequently unavailable and it could hurt your business, especially if your revenue is generated directly from your website (for example if you have an online store.) On the other hand if the site is merely a displays your name and email address, it might not worth an extra $1000 a month to go from 99.9 uptime to 99.999 uptime.

4. Update your contents

People are more inclined to visit your website often if the content is always fresh and changing. The more they come to your website the higher the likelihood of transacting with. Also search engines (like Google) seem to rank you higher if you have a lot of visitors. This means when someone is "searching" for you (using Google, for example) your website will be among the first few that is shown and therefore more likely for new prospective customers to visit your website.

5. Consider advertising

It is not inexpensive but it could pay off if you have the right products or services. There are some simple solutions to get started. pay-You can check out Google's Adwords as a starting point. It would charge you by clicks and/or impressions so in way a the cost is tied to benefits. You can also try direct email marketing.

6. Connect with your customers

Many small business use blogs and podcasts to broadcast contents to their existing and potential customers. You can influence them and they can see you as a subject matter expert when you share your knowledge, vision, and plans. You might feel you are giving away free information but the results often outweigh any perceived loss.

7. Collect data, analyze, and adjust

You can track how many visits your website gets, who they are, which page the start with, how much time they spend, which page they departure from, and many other useful facts. Such information can help you decide why people are attracted to your page and what they want, which in turn can help your maximize what you want to get out of your website.

8. Give customers what they want

It is important that the information on your website is easily discoverable, accessible, and usable. Pictures and diagrams are often better received than words. Also the main points should near the top or even in the title instead of several pages into the article.

9. Have a clear idea and implement it

There should be an objective behind everything on your website. If you are selling products online, help them easily find the product they are looking for, describe clearly (including pictures if applicable), and make the navigation to the order processing page seamless. On the other hand if the website is simply a "business card" put all the contact information on the first page in a clearly visible area such as the top 25% of the page.

New Business Cards

noreply@blogger.com (Kaamel Kermaani) — Sat, 22 Aug 2009 18:54:00 +0000

A few years ago while attending a convention I noticed this self claimed consultant who was handing out his business card. I am using the word card rather generously here as they were clearly cut out of regular paper. Seemingly he had printed them on paper using a laser printer. It looked unprofessional, yet it was thought provoking. Why one needs a business card anyway and what do we do with them after getting them? In the old days when I used to have a note book for taking notes I taped them onto the page where I jotted down any notes from our meeting. In most cases I never went back to the cards. The phone numbers of the few whom I called back were transferred to my phone book (and later my electronic PDA or computer.) The rest were trashed or forgotten for all practical purposes.

Things have changed but not necessarily the principles. I only use paper and pencil to take notes when I don't happen to have my laptop with me. When I get a business card, I usually use it to read the name to make sure I got it correctly, and may be as a reference for the rest of the meeting. As for the contact information, I try to do an email exchange so we both can transfer the info (and in a way to verify the email address). From there we can move the relevant pieces easily to our electronic phone books. This works for me, although it is not as efficient as it can be. In addition, I wish there was a picture to go with the contact info. It is embarrassing when I meet with someone who I had met a couple of months earlier but cannot remember which one of the eight people whom I met with on that day he/she is.

Here is what I think would help. The new business cards should be similar the traditional ones on the front (kind for backward compatibility)! On the back, there should be a photo, plus a barcode with everything on the front side encoded in it. The recipient would use their smart phone to take a picture of the back of the card which then would be decoded and stored in the phone. From there the information can be transferred to (or sync'ed with) the contacts database on one's computer. Wouldn't that be neat? I would think so, but is it practical?

Well, there are some solutions out there already. For example Microsoft has Microsoft Tag, an online application where it can create a barcode image from the provided information. Microsoft also has free applications for smart phones, including iPhone, BlackBerry, Nokia, and RIMM that can take a picture from the barcode and decode it. It worked very nicely when I tried it. By the way, the barcode generator is not limited to contact info. It can be used to encode free text, web pages (URL), and phone numbers. There is even a "custom" barcode option where one can convert the barcode into something that is more eyes pleasing. Microsoft Tag is still in beta, although it appeared to be useful even now.

I am not sure how long it will take for this idea to become common place, but sooner will be better than later. I for one will have my next business card with a photo and a barcode. Would you please do the same too? Oh, as for that consultant, I don't think this will help him, but I still think he should have spent a few bucks on normal business cards.

And that is how I see it …

I am a netbook - I am a notebook

noreply@blogger.com (Kaamel Kermaani) — Mon, 17 Aug 2009 19:31:00 +0000

Have you seen those Mac commercials? The question that I am asked these days more than Mac versus PC is about netbooks and notebooks.

With the integration advancements the notebooks became more desirable than the desktops since they were portable and could do everything that the desktops did a year or two earlier. The desktops became mostly specialty computers where typically higher end processing, and graphics, large amount of memory and disk space, larger (or multiple) screens, or some combination of these were required. And then came along the netbooks to further enhance 3 aspects of the notebooks: the size and portability, the battery life, and the price.

A typical netbook is measured less than 11 inches diagonally compared to more than 17 for some notebooks and weighs about 2 pounds compared to about 7 pounds for an average notebook. This is a huge improvement for those who lug around their laptops all day. And to save space, most (or maybe even all) of these netbooks use solid state drives instead of mechanical drives which also weight less, are less noisy, and don't break (as easily). But the smaller size is one of the most significant drawbacks too. The keyboard is too small, for writing large documents and the screen size might be too small to be comfortable as the primary display. and the flash drives means less disk drive space which could become an issue for those who need to have a lot of files on their systems.

The battery charge might last up to 8 hours for some of the netbooks versus 2-3 hours for a typical notebook. For someone who is out of the office for most of the day this might a deal maker as a single charge could last the entire day. The longer lasting battery is made possible by using lower power processors (and other low power consumption components) which achieve the lower power by reduce processing power. For some applications this might not be a big issue while for others it will be.

However, the last enhancement, the price, is the number one reason that attracts people to the netbooks in my experience. They can be had for anywhere between (supposedly) free to about $300 in most cases, compared to $600-$3000 for notebooks. The most intriguing the least understood part of the very low end of the price range is that part (or even all) of the actual cost is subsidized by the cell phone providers almost exactly the same way they subsidize the cost of the cell phones. That means a two year (in almost all cases) contract to a wireless data plan for around $60 a month, which if you do math will cost $1,440. If one needs the data plan (and some, but not all, do) then it is OK. Otherwise this has be calculated into the total cost of ownership. Additionally, the lower cost is achieved by eliminating or reducing some features (graphics acceleration, processor, memory size, disk size, IO connectors, etc.)

So there is no clear "winner" here, at this time. Many use their computers primarily to access internet and a netbook might be a perfect solution for them (that is why it is called NETbook.) On the other hand some do a lot of intensive work on their laptops (spreadsheets, heavy document editing with graphics, charts etc, and other similar applications), needing large keyboard and screen size and lots of memory and disk. For them a notebook is the right answers. Which one is right for you? Only you know the answer.

Down the road, the netbooks will displace the notebooks in much the same way that the notebooks have been replacing the desktops, although they might not be call netbooks. We will be able to pack more functionality at a lower price into lighter laptops with larger screen sizes and keyboards, similar to the Apple MacBook Air, but at a much lower cost (closer to $300 than $3,000.) They In the meanwhile, the competition between the netbooks and notebooks is bring down the price of all laptops, which is good news for the rest of us.

And that is how I see it ...

BodyBugg: SaaS to Lose Weight

noreply@blogger.com (Kaamel Kermaani) — Mon, 10 Aug 2009 21:49:00 +0000

There is a product called BodyBugg which is featured on the TV show “The Biggest Loser”. For those who are not familiar, it is a little gadget that one wears on the back of their upper arm, throughout the day. It has some sensors and a very accurate pedometer. The sensors measure things like body temperature, the amount of sweat, and “heat flux” (whatever that means) and the pedometer counts the number of steps one walks. Using all that information, the company behind this technology, BodyMedia, can calculate the number of calories burned, or that is what they claim.

Anyone who has ever tried to control their weight can attest that such a tool could be invaluable. BodyBugg does not directly show how much is burned. It only collects and stores up to several days worth of data internally. Using a personal computer the data will need to be downloaded to the BodyMedia servers via a Java application that runs in a browser. By the way their website can track what one eats, which is then converted to calories too. In addition it can track body parameters (weight and various body measurements). And if that is not enough, depending on the goals, the websites puts together a suggested daily meal plans for breakfast, lunch, dinner, and snacks, as well as daily exercise routines. Overall the combination of BodyBugg and this website is a very powerful fitness tool.

When I started using BodyBugg which I had received as a gift last Christmas, I noticed my calories burned at night when I was sleep was very close what one would expect by just calculating the resting metabolism rate (RMR) for an average person in my situation. I always had experienced a higher calorie burned rate compared to everyone else around me and couldn’t agree with seeing an “average” metabolism. As I read more about the device I started picking up clues as how their website was calculating the results. Based on that I modified some of my inputs (age, weight, etc) to achieve the expected RMR (more on this in a later blog, maybe). After the adjustments I thought it did a reasonable job of helping me track my calories. After several months of usage, my weight was relatively close to what the tool was predicting so I was fairly happy.

There were a few things that I don't like about it though. First, it does not show all the statistical data that I am interested in and there is no (known) way of downloading the data into a spreadsheet that I can use to do my own analysis. Also, I had to fake some data to make it work better. I would have preferred if it could allow me to enter the RMR directly (one can measure their RMR in a clinic) instead of them using an aver RMR based on age, height, and weight. Lastly, if their website is down (which happens a couple of times a month) one cannot use it at all. In other words there is offline solution. This can be problematic. Lastly, there is no way to enhance what they offer any further as there is no way to interoperate with other online or installed fitness tools. Whatever they offer is all get and there isn't much else you can do.

So, while I like what they have done, they need to do more. There needs to be a solution for data portability. They need an offline application that at a very least can function for a few days before needing to interact with their servers. I wish they open up their APIs. That could allow other vendors (or even the end users) to use the collected data to provide enhancements and functionality beyond the current scope. And lastly, interoperability with other tools like heart rate monitors and GPS can take really BodyBugg to a different level.

Do you Google Voice?

noreply@blogger.com (Kaamel Kermaani) — Sun, 09 Aug 2009 20:32:00 +0000

First things first, if you haven’t already, go to http://google.com/voice and sign up to get your Google Voice number. In a week or two you will get an invitation. You can then claim a free GV phone number, not a phone line, just a phone number.

What can you do with it? You can set up the Google Voice to route the calls to your real phone lines like cell phone, home, office, or whatever. You can even set it up to call different numbers at different times of day or different days. It also comes with a voicemail that can be set up with different greetings for different callers. And to top it off, it can transcribe the voicemails which can be read on a computer, emailed, or saved away. And then there is the free text messages and lower international calling rates (calls from the US to the UK are only 2c a minute as an example.) To use Google Voice to call out, one would enter the number on their website or directly on the cell phone if it is running the GV application. Calls to any number within the US are free. The app on the cell phones make the experience seamless. (See my previous blog about the app for iPhone.)
I like GV for several reasons. I have started thinking of it as my permanent phone number. When any of other numbers changed, all I would need to do is to redirect the Google Voice number to my new number(s) transparently. Another benefit is that if I am somewhere with poor cell coverage where, I can have my GV reach me a landline number while I am there. I can also have my calls go to my cell phone and office number during work hours, both home and cell phone during my commute hours, and only home, later in the evening. Another side benefit is that I might be able to “read” my voicemail while “attending” a boring meeting.
Once I had my GV, I realized that I actually needed (or wanted) 3 GV numbers, one for my cell, one for home, and one for work. You see, I don’t prefer to share a single number for everyone who might need to reach me. There are times that I would want calls from everyone to go to the same number or some combination of home, cell, and office numbers. But there are other times that I would prefer different calls to be routed to a different numbers. As an example I don’t want my office calls to reach me over the weekend or after hours, and I don’t want my home calls to reach me in the office (most of the time).
So I used three different email addresses and got 3 different numbers. And that is when I noticed a couple of flaws with GV.
First, GV refuses to allow more than one GV number to be routed to a cell phone number. I am not sure why, but that is how it is. It allows up to two different GV numbers to be routed to any phone number that is not a cell phone (this can be bypassed to claiming the line as landline, but then there would be no text messages for the line). Both of these are somewhat serious limitations in my opinion. I can see a not everyone in a family of 4 (and nowadays that means 4 cell phones) being able to route their GV calls to the home number. And if there is a shared office number (in a small office setup for example, or when a single admin supports multiple managers) only two of the employees get to route their calls to that line. And of course if there is a shared cell phone then all users but one would be out of luck.
Also, a couple of Blackberry users have complained about using the SMS feature. In general, the incoming text messages are kind of weird as they show up coming from a strange number, although it kind of clarifies the actual senders, but still it is not clean at all. I also I hear that the application on BB is not flawless yet (call connections are slow and at times don’t go through.) But I cannot be sure if those are the app’s problem or GV, which by the way is in Beta at this point. The thing I hear is that the transcriber is good but not perfect.
So, to sum it up, I would say it is an awesome service, especially since it is free. But I also must add that like practically every other service that Google offers, there are some fundamental architectural issues. The limitation on the number of phone lines is artificial, and reminds me of the current issue with the Contacts in Gmail which don’t sync with Outlook, for whatever reason. And the problem with the text messages reminds me of the “on behalf of” problem that Gmail had for a long time until it was fixed very recently. I am not sure why, but Google seems to consistently go a long way in providing almost perfect services, only to leave a couple of key features out as if they want to make sure their services don’t dominate.

Google Voice App for iPhone: Apple and AT&T

noreply@blogger.com (Kaamel Kermaani) — Thu, 06 Aug 2009 21:27:00 +0000

Last week FCC asked Apple and AT&T why the Google Voice application for iPhone was rejected by Apple (read more here). Google Voice is a new service from Google (by invitation for now) where they give users a free virtual phone number that can redirect calls to users real phone numbers. More on Google Voice and why this is good and useful later.
Up until recently there were a couple of 3rd party apps for Google Voice for iPhone on iTunes store, which the sole mechanism to distribute apps for iPhone. Google also developed an app of their own and submitted it to Apple to be publishing on iTunes store. Amazingly enough Apple rejected Google's app, AND they even removed the previously published 3rd party Google Voice apps from iTunes! The reason given by Apple was that the apps duplicated the functionality already present on the phone!
Now FCC wants to know why the apps were rejected and removed, and also if AT&T had anything to do with it. Clearly AT&T cannot be too excited about this service. For one offers free SMS, and it also offers low cost international calls. So, Google Voice presents competition and not surprisingly AT&T does not like it. This is not to say that AT&T is the one who asked Apple to reject/remove the apps, but clearly one can see the conflict of interest. It is hard to find a logical reason for Apple on their own to reject the apps as they didn't appear to cause any technical issues, or otherwise cause liability or concern of any kind. Their presence seemed to be inconsequence to Apple (other than potentially at odds with the terms of their contract with AT&T.) At any event, the stated reasoning might, in my opinion, be construed as anticompetitive and conceivably illegal, at least in some jurisdiction.
When MS introduced the Internet Explorer and integrated it into their OS, they were prosecuted for anti competitiveness and lost the case in Europe (many argued they got away in the US). It would have been very interesting if they had not allowed other browsers to run on Windows by claiming "similar functionality already exists" as an excuse. But Apple feels this is not applicable to them. Given the number of independent application developers, and the openness of the iPhone OS, and the number of units sold, Apple is now where exactly MS was back in those days. Even if AT&T has nothing to do with it, Apple should not be allowed to use this kind of logic. And if it turns out that indeed AT&T is behind it in some way (I must admit I cannot resist thinking that they are), then this is clearly a collusion, and in my opinion meaningful restrictions should be imposed on both to curtail future conspiracies and to protect consumer rights.
Frankly, I don't see any reason for Apple to have any authority to prevent applications from being published, unless they cause technical issues that are harmful to the device in ways that implicate Apple. AT&T also should be prohibited from interfering with the app publishing process, except for the cases where the apps cause technical issues that affect to their delivery of their services or use their services over and above what the customers have already paid to use. Loss of revenue to competition for either Apple or AT&T should not be a consideration.
And that is how I see it ...

Related stories:

DSL or Cable?

noreply@blogger.com (Kaamel Kermaani) — Mon, 03 Aug 2009 18:09:00 +0000

I have been asked this question many times. I recently posted my opinion on Linkedin and wanted to share it here too.

Assuming both are available (in many areas in the US you get one or the other), then it boils down to price, performance, and reliability. This last one is a vendor thing and not (necessarily) a technology thing.

The state throughput of DSL (in most cases) is actually achievable while the cable bandwidth is shared with your neighborhood and seldom is realized, certainly not at the peak hours.

For both, one would usually (in the US) get a lower price for the first year, and assuming both are equal from the performance and reliability point of view it might be advantageous to switch annually, or at least threaten to switch, to get the best price. Generally the prices are very comparable. But sometimes cable companies bundle in some TV services for a lower total cost.

As for the performance, it depends on where you live. In my area Comcast (cable) and AT&T U-verse (DSL) are both available and cost about the same. Comcast offers 16 Mb/s while AT&T offers 18 Mb/s. I used to have Comcast (when they were faster than AT&T) but their reliability was unreal (about 10 hours of downtime/month on average). I had measure 6-12 Mb/s (maybe even 16 Mb/s in rare occasions) in actual transfers from faster sites. I switched to U-verse about 6 months ago. It's reliability is solid (I have not noticed any downtime since I started the service) and its performance is very predictable and very good. I have even measured faster than the advertised speed of 18 Mb/s, but routinely get over 16 Mb/s (almost always limited by the site) when download from faster sites.

So in my experience it is a no-brainer: AT&T U-verse (DSL) hands down.

My Health Care Reform Proposal

noreply@blogger.com (Kaamel Kermaani) — Sat, 01 Aug 2009 22:15:00 +0000

There are so many issues with our health care system that no matter where we start it should not be too difficult to improve on it. Yet there are so many interest groups that are living off the inefficiencies of the current system that any change will be resisted vigorously. So I am going to ignore all the secondary issues, the ones created by those groups, and go right to what I would do.

Problems have to be addressed from the top, and that means the insurance companies. They are feeding money to the rest of the health care system and where ever they go, the rest will have to follow. As a starting point I would remove any medical history, age, geographical location, or anything else from the basis of insurability or pricing consideration. In other words they must offer insurance to everyone, and at the same price. People could choose the insurance company and sign up by providing their name and sending in the payments. Some will argue that the insurance cost for the healthy and young will go up, and in effect they would be financing the old and the unhealthy. The fact of a matter is, everyone will eventually get old and unhealthy, so everyone would be paying the life time average cost, for themselves. The competition would in the long run, if not immediately, bring the insurance prices down as all the insurance companies would have no choice but to take their total cost and average it, and then compete with each other on that average. This is currently happing only for the youngest and the healthiest, and those are the ones that might actually not need insurance as much as the rest.

Then I would propose a basic national insurance that everyone would receive. Something that would cost like $10 per visit for the first 5 visits in a calendar year and then %20 deductable, not to exceed a certain amount of out of pocket per year. All insurance companies would be encouraged to bid on this national plan with no preset limit. The lowest cost bidder would win and get to be the national insurer, and no matter how you cut it, make a killing. The competition would be fierce which could make the bids very attractive. To make it fairer, there might be schemes like up to 5 would win as long as a) they are within the 10% of the winner, and b) they agree to accept the winner's bid, or some other scheme. This would help having multiple national insurance companies.
The insurance companies can also offer other plans to general public, for example lower deductable or lower percentage, or coverage for some semi-essential procedures (eye care or dental, for example). People would be allowed to opt out of the national coverage and go directly with the insurance companies, but would get tax credit equal to what the base national cover would cost.

The cost of the national insurance would come from higher taxes from the employers (not employees). No one likes paying for anything but we all want to be rich and beautiful. In this case the employer would pay taxes instead of paying for employee health care insurance. It should not be a huge burden. I would argue that the cost of insurance will go down and as a result employers in the long run could be even paying less. But even if they were not, this plan would guarantee insurance for all of us. And it would be a lifelong insurance coverage independent of health, age, or employment status. This would even render other forms of national insurance like Medicare, etc unnecessary and therefore make this plan less costly.

I must be sick ...

noreply@blogger.com (Kaamel Kermaani) — Sat, 01 Aug 2009 21:08:00 +0000

I was never called overweight, fat, or anything like that. But as I was getting older I noticed that my pants size was changing. In my latest upsizing, I was wearing size 34, and admittedly my belt was not exactly over my tummy, rather kind of below the widest part of midsection. Otherwise I needed size 36. And to make it worse, it seemed like the increase was accelerating. Finally, one morning I decide to sep on the scale just to check things out. We have one of these gadgets that shows the weight and the body fat percentage, although I don't think the fat percentage is very accurate. It showed 165 pounds and 25% fat. It was not terrible for someone who is 5'11", but I was inching into the overweight zone.

I decided enough is enough. First I started reading a little about weight loss. My daughter already was ahead of my by a couple of years (she actually was/is skinny, but you know how girls feel about their weight.) She helped get started. It became clear that I was not active enough. I learned that most doctors recommend walking about 10,000 steps a day. I bought a pedometer (actually I tried several before settling in) and noticed that I was barely getting to 1,500 on an average day. The solution was easy: walk more. I tried parking the car a little farther and stopped looking for shortcuts on my day to day chores. Soon I added a daily routine of walking after lunch and then later after work. It took a while but I did reach 10,000 steps. The problem was that it was taking too much time.

The next move was to start going to the gym instead of walking and seemed to be a little more efficient. Then I did the gym AND the daily walks. I changed my diet too and started watching what I ate. Labels are great and tell you everything you need to know. My weight was clearly dropping by all measures, including my pants size. My weight reached 160, then 155, and continued to go down. I was wearing my older size 32 pants. It took a while but I was actually wearing my belt right where it was supposed to be.

I started hearing comments and questions about my health. "Are you OK? You seem to be losing weight." "Have you seen a doctor, you look weak." "How is life treating you, are you stressed out?" It was funny, but there were so many doctors around me who could diagnose my "disease" by just looking at me. As I explained the situation, the comments slowed down or subsided all together. Shortly after I was being asked about the details of what I was doing, as there seemed to be an interest. I noticed that the number of people who were wanting to lose weight was rather large. At the same time most were not impressed by my routine. They wanted another solution that was so rigid and painful. More like taking a pill in the morning or just before going to bed. They wanted to heard that it is not what you eat, but when you eat it, or how to eat, or what else you eat to compensate. Many decided that it is genetic and nothing would matter anyway.

By the time I reached my goal, I was going to the gym 3 times a week and walking for over an hour every day. Because I was fairly active I could eat a lot of food and still not gain any weight. I liked it a lot as I really enjoy eating. Those who watched me eat like a pig and remain "sickeningly" skinny now were commenting that I "really" had to see a doctor as my body was not absorbing the food . Which by the way I need to do for my annual checkup. By the way, this morning the scale showed 140 pounds and just 7.5% body fat. My pants size? I am wearing size 30 and that is because I cannot find any size 29 pants that is 33" long. It is a compromise that I have to make.

I must be sick ... And loving it.

Who is getting the upper hand in the Microsoft-Yahoo deal?

noreply@blogger.com (Kaamel Kermaani) — Fri, 31 Jul 2009 22:46:00 +0000

According to the published reports Microsoft and Yahoo agreed to a 10 year partnership, subject to regulatory approvals. Yahoo will drive the sales and marketing arm while Microsoft will be the technology driver with their new search engine: Bing.

Reportedly Microsoft will give Yahoo 88 percent of the search ad revenue made on its Web site over the next 5 years. In fact for the first two years, Microsoft will pay 110%; that is an extra 10% out of their pocket.

So who wins? Microsoft was willing to pay a "boatload of money" to buy Yahoo not too long ago. With this new arrangement, they are only paying Yahoo 88% of the ad sales over 5 years. Being the technology provider, Microsoft's search engine (Bing) will become the real IP and the Golden Goose in the partnership. In effect Microsoft is buying the second position in the search engine market (after Google) with no cash up front. Another way to look at it is that they are changing their original Yahoo acquisition plan to a try-and-buy. If things don't work out, they will keep the money they were planning to spend buying Yahoo. If they do, they will own the market share they were eying from the beginning.

As for Yahoo, they will be financially set for the next two to five years. This will give their management a chance to work on other technologies and solutions. If they the right vision and can come up with the right strategy and put together teams that can execute, they could create new business models that would work with or without Microsoft search engine. If that happens they will survive, and might be in a much stronger position than they are today.

Oracle-Sun: What is next?

noreply@blogger.com (Kaamel Kermaani) — Wed, 29 Jul 2009 20:49:00 +0000

It will be interesting to see what Oracle will do with Sun once it takes over their day to day business. I see 4 different pieces:

· Solaris, Java, MySQL
· Storage business, the old StorageTek
· The x86 server business
· The SPARC server business

Oracle was reportedly only interested in the first piece, initially. My thinking is that MySQL will somehow be integrated into the overall database business. Java is of value to them both internally and as a licensable product. Solaris can become the primary operating system for their products. I can see them optimize their products for Solaris, making it a very attractive technology for their customers. They might actually market it as a viable alternative to Windows Server, going head to head against Microsoft.

The storage business is different enough that I don’t think they prefer to keep. It could be a distraction and there is little synergy with their existing products. My guess is that they will just sell it to the highest bidder or spin it off for now and sell it later.

The x86 server business will be intriguing. With minor modifications to HW and Solaris-x86 they can offer an all-in-one platform (like Mac and OS X). They can then optimize their products for that platform creating a unique package with practically no competition. There would be minimal overhead in managing an x86 product line as it is a commodity hardware that can be outsourced easily. On the other hand this can annoy the other HW vendors like IBM, HP, and Dell. The might not not be a good thing for them in the long run.

The SPARC server business in my view could have been the least desirable part this acquisition. The processors are mostly developed internally (by Sun) and it will take a ton of cash to continue the R&D. In addition the growth has been stagnant and would have likely been on the decline on its own, but could get worse if there is any perception of uncertainty. I would think Oracle will try to sell the SPARC business but there might not be many takers. The likely buyers should be IBM and HP. They would consider it only as a means of getting Sun’s customer and migrating them to their own product lines. There is at least one other viable buyer that I see, Fujitsu. They currently supply the high end servers to Sun and are the only other company who has been investing in SPARC development. Under this scenario SPARC might live for a while, or maybe even for a long while, as Fujitsu has the expertise to keep that technology advancing. As the sole player in the SPARC market they could make enough money to make it a real business. That could be a win-win-win, for Fujitsu, for Oracle, and for the end users.

Cloudy days ahead?

noreply@blogger.com (Kaamel Kermaani) — Tue, 28 Jul 2009 06:51:00 +0000

I feel it is more of an evolutionary step and therefore might have a slow adoption rate. New non-critical applications, the ones mostly used in schools and by individuals will help hammer out security, reliability, portability, and availability, and will also help establish the baseline for the overall value. As the more natural and more critical business applications penetrate into the CC, the acceptance will accelerate. As the more serious issues are addressed over time, the breadth of applications that can naturally move to the "cloud" will also increase.

In my opinion the six fundamental issues and obstacles that will need attention are:

Availability - the data and the applications should remain available when needed, even when there is no connectivity.
Performance - for the end users the perceived performance of CC should be superior or at least similar to using a laptop.
Security - The data needs to be protected and secure.
Portability - If the customers is not satisfied with a CC vendor (or in the worst case if the vendor goes out of business) the data and applications should be seamlessly portable to another vendor.
Low cost - the cloud route should be less expensive than buying a laptop and all the necessary applications, plus maintenance cost. The less complexity is good, but should represent one of the benefits and nothing more.
Device independence - the data and the applications should not tied to a particular HW or system.

Google, Microsoft, salesforce.com, and others are already moving in this direction, although some are ahead of others. Storage in the cloud seems to be the leading application, followed by emails systems, and other similar online services. Microsoft has announced that Office 2010 will have a free online version. On the server side it is natural to run the database servers remotely, if the data can be moved quickly between the office and the datacenter. Desktop virtualization is another element that can connect the users to datacenters. Many other pieces are already in place or in the works.

The solution providers who can pull all these together and put together usable and simple solutions will get ahead. Who will they be? Time will tell. But if things are not worked out rapidly, there might be other technologies coming along and outdating CC before its times.