Tom Petrocelli's take on technology. Tom is the author of the book "Data Protection and Information Lifecycle Management" and a natural technology curmudgeon. This blog represents only my own views and not those of my employer, Enterprise Strategy Group. Frankly, mine are more amusing.
Thursday, October 20, 2005
The new "toy" that I've been playing with is called SyncToy. It's one of the many Power Toys - small but useful utilities - that Microsoft comes up with from time to time. Some of their Power Toys are okay, some a waste of bits, and others genuninely useful. SyncToy is in the last category. It's very useful indeed.
It is simple in concept. SyncToy allows you to choose two folders, aptly called a "Folder Pair", and synchronize them. There is the usual basket of options. Completely synchronize both folders, only copy from one folder to another, that sort of thing. You can also have many different folder pairs and choose which ones to run all together and which ones to run manually.
So what's the big deal? It's a neat way to back up your critical desktop files to a network drive. One of the big reasons that people lose data such as word processor files is because they don't copy them to a network drive where the usual backup system can take over. SyncToy automates this process, makign it quite painless. This is especially important for people who work in a SOHO environment which does not have hefty IT resources.
Now, you can also set up SyncToy to run automatically. You can use the Windows Scheduled Tasks feature to launch SyncToy at a specific time (say, the end of the day). If you use the -R switch on the SyncToy command line, it will automatically run whatever group of Folder Pairs that you have configured to run. I've set up scheduled tasks to run SyncToy every day at the end of the day and sync my critical files to my network drive. Instant backups!
So, if you want a toy to play with, try SyncToy. It's not as much fun as Doom 3 but its way more useful.
Friday, October 14, 2005
There was a lot of speculation that it would allow eBay to connect buyers and sellers using voice. This way , if you had a question about an item and the seller was around, you could just phone them. Wow! That must be it.
Not! The whole point of eBay is that you don't have to have long conversations to buy and sell things. Can you imagine what it would be like to be an eBay seller and have every moron in town who didn't understand what you were selling, or simply wanted to haggle around the auction system, bugging you on the phone? It would drive sellers away in droves.
Even as a buyer, I don't see much to like about this. So I can ask a seller questions. That only gives them an opportunity to talk around problems with their products. If a seller is going to yank my chain they will do it more easily on the phone then via e-mail, where it's written down. Again, where's the advantage.
I think that the advantage is really in the other direction. As someone who has used Skype for some time now, especially for business purposes, I can attest that it works and is really useful. It's free as long as you stay in the network. If you want to call a regular old PSTN phone, then you have to pay a fee. Same goes if you want the ability to have old-fashioned phones call you. You buy minutes, just like a cell phone.
There's the clue. Mobile phones suffer from the need to predict your usage in advance. No ones usage is ever that steady. Sometimes, you want to talk a lot. Other times, very little. So you either end up paying overage fees or not using the minutes you have already paid for.
Now, imagine for a minute, if you could sell your unused minutes to someone who needs them now. Later on, you could buy more minutes from someone who has extra that they aren't using. How would you price these minutes? Should you price them higher because it's the holidays and you know people want to talk? Price them lower because you don't know what the demand is? Too high, and no one buys them. Too low and you leave money on the table and feel like an idiot. What if you let the market decide? What if you instead open the sale up to bidding? Like on eBay.
If you think this is far fetched, please be reminded that telecommunications companies do this all the time. They buy and sell bandwidth (and minutes are a form of bandwidth) on exchange markets. Electricity is sold this way too. The only difference is that the big boys do it in massive quantaties, in closed markets, that are hard to join. They aren't interested in your 50 extra minutes not do they really want you to try and sell them. They especially don't want someone to buy them and not pay overage fees.
Skype and eBay have a different perspective on markets. Skype offers basic services for free and would love to have us all buy Skype minutes. Since you don't lose them like most cell phone plans, they really don't care who uses the minutes so long as people buy them in the first place.
eBay, loves small transactions. That's what they do.They enable regular joes and jills to sell anything no mater how small or odd, so long as it's legal. As long as activity is generated, they make money.
So, it is in the best interest of both companies to create a communications system where a small number of minutes can be purchased, then resold many times over. A combined eBay/Skype makes money selling minutes and again when consumers buy and sell them.
For consumers, this could be a true breakthrough in telecommunications - The ability to make maximum use of the resources we purchase. I don't doubt that it will be only a matter of time before other premium minutes become trasnferable and hence open for sale.
I'm sure there is tremendous risk to eBay in this acquisition. It's expensive and based on an unproven busienss model. But if it works, it will transform telecommunications forever.
Friday, October 07, 2005
The problem comes when you have Linux-based systems. Don't get me wrong - I like Linux. I've had a Linux system running somewhere for ten years. Software vendors appear to like Linux too. Almost every major data protection software company has support for Linux - or so it would seem.
What vendors support is specific, major, distributions. It's no different then their Windows and UNIX support. They can't engineer solutions for all flavors so they pick what they think will be the most common ones or the ones where they can get support from the manufacturer.
That makes it tough for Linux users because not all Linux distros are released by companies or organizations with a lot of market presence or support staff. From the software vendor's perspective, that distro is not worthwhile. Support for major companies' Linux products are certainly available. If you are a Red Hat or Suse/Novell user, you're usually in luck. If Debian, Knoppix, Slackware, or some other distro is your favorite flavor, your chances are much less that you will find decent support for advanced data protection features.
To get an in-your-face example of this, look at the offerings from Symantec (what used to be Veritas). We are talking about one of the most important data protection software companies in the industry. Linux Support is only for SUSE and Red Hat. There is no application specific supprot for any of the common open source databases used with Linux. No MySQL support and no PostGresql support.
And if you think the Open Source community will ride to the rescue, think again. Although the Open Source database folks are putting features like replication in their databases, the tools available from the Linux community are pretty scarce. The reason for this, is that this stuff is very hard to do. It's not for the weak. A bunch of people writing code on weekends isn't going to come up with good data protection software. A lot of specific skills are needed and the people with those skill are working for Microsoft, IBM, or Symantec.
So folks. If you are serious about Linux, you need to stick with SUSE and Red Hat. Otherwise, you simply won't have the tools needed to keep your data safe.