Q: Build it or buy it?
A: Depends how good you are.
When asked to solve a problem, just about every software developer in the known world will create a solution, maybe one that is cost effective and fast to deliver, maybe one that is expensive as all hell and prone to more missed dates than Windows Vista. On the other hand, ask a business manager to solve a problem, he or she will hit Google, find a purchase to meet the need, and if The Budget allows, cut a purchase order.
Engineers want to make stuff. Business managers want to make money. That’s the way it is in our well ordered Universe.
Up until today I was in the “buy” camp, which is pretty funny given that when push comes to shove I define myself as a software developer. I mean, the consolidation of the software development industry into low cost, off shore coding centers in India, China and Eastern Europe has made it almost impractical to do any sort of profitable commercial development for the general market. If you do want to make a custom solution, most times it pays to buy some off the shelf components and just stitch them into your code. Or you can forget stitching anything together. Just use one of the big ass portal products such as Microsoft’s SharePoint, the open source DotNetNuke, Joomla for PHP, or WebSphere’s Web Portal and configure it to be the app that you want.
Now don’t get me wrong, I love building software, in fact I love building anything. I am a builder sort of guy by nature. But, I can see the handwriting on the wall. Making software is hard and expensive. Most custom software projects that start never see the light of day, let alone become a reliable mainstay of the organization that plunked down the bundle of money required to get it out the door. If some company has the brains, tenacity and deep pockets that are required to actually make commercial software that works, well far be it from me to let my ego get in the way. I’d just rather buy something that I know works and learn how to use it. I mean if, I really gotta code, I can just make something and post it on SourceForge or space that I rent from GoDaddy.
As I said, my brain says buy while my heart says build. Then I read this month’s issue of Wired. It seems that Google is building a Big Ass Data Center out in Oregon. The company is building in Oregon because there is plenty of cheap power to be had coming off of the waters of the Columbia River as it roars through the turbines of the Dalles Dam. The undertaking is stupendous. We’re talking about tens of thousands of servers under one roof chugging away as one big, RAM based collective intelligence processing millions of queries a second worldwide. This is not an undertaking for amateurs. We’re talking a data center that will consume as much power as Las Vegas.
They’re building it! Not buying it from Verison or ATT.
But, then again, Google has been building it’s infrastructure from the get go. The company figured out a while ago that it could take some off the shelf processors, power supplies, RAM and hard drives and turn them into world class server farms doing things that used to be the province of mainframes and supercomputers. Google builds because the company has the competence to build. In fact, I believe that any organization will build if it has the competence to build. But, when the competence is gone, what’s left?
So now the question I have to ask myself is this: If Google has the competence to build big stuff, what’s stopping the rest of us who are just trying to build the small stuff? We just can’t seem to be able to ship. Granted, component based software is the way to go. No sense reinventing the faucet every time we need to build a bathroom. But jeepers, some days it seems as if we can’t even build the sink.
So, if you can’t build, you buy. And, if you can buy…. well, you sell.
Maybe that’s why there seems to be so many mortgage brokers around.
When asked to solve a problem, just about every software developer in the known world will create a solution, maybe one that is cost effective and fast to deliver, maybe one that is expensive as all hell and prone to more missed dates than Windows Vista. On the other hand, ask a business manager to solve a problem, he or she will hit Google, find a purchase to meet the need, and if The Budget allows, cut a purchase order.
Engineers want to make stuff. Business managers want to make money. That’s the way it is in our well ordered Universe.
Up until today I was in the “buy” camp, which is pretty funny given that when push comes to shove I define myself as a software developer. I mean, the consolidation of the software development industry into low cost, off shore coding centers in India, China and Eastern Europe has made it almost impractical to do any sort of profitable commercial development for the general market. If you do want to make a custom solution, most times it pays to buy some off the shelf components and just stitch them into your code. Or you can forget stitching anything together. Just use one of the big ass portal products such as Microsoft’s SharePoint, the open source DotNetNuke, Joomla for PHP, or WebSphere’s Web Portal and configure it to be the app that you want.
Now don’t get me wrong, I love building software, in fact I love building anything. I am a builder sort of guy by nature. But, I can see the handwriting on the wall. Making software is hard and expensive. Most custom software projects that start never see the light of day, let alone become a reliable mainstay of the organization that plunked down the bundle of money required to get it out the door. If some company has the brains, tenacity and deep pockets that are required to actually make commercial software that works, well far be it from me to let my ego get in the way. I’d just rather buy something that I know works and learn how to use it. I mean if, I really gotta code, I can just make something and post it on SourceForge or space that I rent from GoDaddy.
As I said, my brain says buy while my heart says build. Then I read this month’s issue of Wired. It seems that Google is building a Big Ass Data Center out in Oregon. The company is building in Oregon because there is plenty of cheap power to be had coming off of the waters of the Columbia River as it roars through the turbines of the Dalles Dam. The undertaking is stupendous. We’re talking about tens of thousands of servers under one roof chugging away as one big, RAM based collective intelligence processing millions of queries a second worldwide. This is not an undertaking for amateurs. We’re talking a data center that will consume as much power as Las Vegas.
They’re building it! Not buying it from Verison or ATT.
But, then again, Google has been building it’s infrastructure from the get go. The company figured out a while ago that it could take some off the shelf processors, power supplies, RAM and hard drives and turn them into world class server farms doing things that used to be the province of mainframes and supercomputers. Google builds because the company has the competence to build. In fact, I believe that any organization will build if it has the competence to build. But, when the competence is gone, what’s left?
So now the question I have to ask myself is this: If Google has the competence to build big stuff, what’s stopping the rest of us who are just trying to build the small stuff? We just can’t seem to be able to ship. Granted, component based software is the way to go. No sense reinventing the faucet every time we need to build a bathroom. But jeepers, some days it seems as if we can’t even build the sink.
So, if you can’t build, you buy. And, if you can buy…. well, you sell.
Maybe that’s why there seems to be so many mortgage brokers around.
posted by Bob Reselman at 7:20 PM
3 Comments:
I'd be careful looking at Google for lessons in software product development for other companies.
What works for Google won't work for you and me, necessarily, unless we're building the next Google.
What they are doing hits the limit of what current software and hardware is capable of, so they have to roll their own.
But for your typical data in, data out business app, you don't need to roll everything yourself.
You only build if you can't find anything other people have done that will work or is too expensive. I've built because either ...
1) I couldn't find anything that other people have done that will work. I know what needs to be done.
2) Other people have done this but to buy their product is too expensive (*cough*HL7).
Have you ever realized that when you buy readymade suits you get a choice of only a few colors & styles, also finding the perfect fit are quite difficult? Wouldn’t it be better to choose from over 2000 different British & Italian fabrics and get a tailor made suit at a similar price that you pay for readymade suits?
We are a company based in Hong Kong and have been providing custom made suits & shirts since 1997. With representatives in major cities around the globe we can arrange to show you the fabric samples and take your measurements, or you can also place your orders online with the help of our measuring guide. There are over 2000 fabrics to choose from along with all the latest styles.
All our suits and shirts are produced by highly skilled Shanghainese tailors in Hong Kong and delivered in about 4 weeks, express delivery can be made in 2 weeks at a minor extra cost. In case you are not able to find what you are looking for then please let us know your requirement may it be in words or by a photograph and we could arrange it for you.
We also have an outlet at the Hotel Intercontinental Budapest where you are most welcome to visit us. Though we are not located in streets like Savile Row (London), we have still been able to offer made to measure suits to many VIP’s from around the world.
Experience an easier way of shopping for bespoke suits & shirts at Euro Tailors
Kenny Surtani
Post a Comment
<< Home