William Adams' WebLog

Esoterica for a new Millennium

Build or Buy?

That is the question.

Be it nobler to pull up your sleeves than to bow to the will of the component provider?

So, here's the deal.  This past weekend, Yasmin and I finished the path around the back yard.  It took 20 bags of bark hauled from the hardware store to the back yard.  Pouring the bags out, and raking them smooth was a much more enjoyable task than hauling the border bricks, or digging  up the grass.

All told, this project cost... I don't know, at least 16 hours of hard manual labor.  About $300 in various materials, and quite a few gallons of sweat and peanut butter sandwhiches.  If I put my time at about $50/hour, that's another $800.  So, a path around the perimeter of my back yard, $1,100.  Path around back yard, $1,100, time with Yasmin... Priceless. 

I could have hired a landscape person to come do the same thing, and it probably would have cost about $400 or so, but hay, it's my design.

Then, the 18th was the nanny's birthday.  Recently Yasmin and I have been taking picutures with the digital camera.  This past weekend we went out to Mt Vernon and took a bunch of pictures of tulips.  Yasmin's actually pretty good at taking pictures.  She's probably better than me, and I attribute it to the fact that she's shorter, so she captures more unique perspectives.

At any rate, we thought we'd use the pictures to make a calendar for the nanny's birthday present.

Build vs buy... Let's see, I can go to Barnes and Noble right now and buy a calendar on sale for about $4.  OK.  So, that's a starting point.

First, we had to buy a printer that could print nice high quality photos.  I have a Canon S50 camera, so I thought I would get a nice Canon printer.  Of course I had to have the one that I could plug the CF card into directly.  the i900D, cost $250.  Then of course the extra paper, ink, and extended warranty... That's another $100 or so, for a grand total of $350.

Alright, now I can print nice pictures, in sizes from 4x6 to 8.5x11.

I need to be able to print monthly calendars.  Luckily, I have Calendar Creator, so I didn't have to write that piece of software, although I could have if need be.  Now there's assembly.

This is no ordinary calendar.  With the high quality pictures that Yasmin took, and the good calendar job, we can just staple the job together like some grade school project.  Nope, this will require one of those nice binding jobs like you get with presentations at work.  OK, off to OfficeMax/Depot/Staple/Supply center.  I just need one of those binders thingies that punches a bunch of holes and helps you put on that plastic spiral thingy.  Of course the $50 one is no longer available, so I have to bump up to the $135 model!  Just put punch uniform holes and put in that platic binder thing.  I am now the proud owner of the GBC Combind C100.

Quick total:  Printer and accessories $350, Binder, $135, 12 images at $.50/ea = $6.  Time, well, it was around midnight when I finished, and I started at around 7:00, so 5 hours == $250.  Grand total... $741

It bit pricey for a calendar, but hay, it's my design darn it!

I know there's a lesson to be learned here about software development, but I'm in sticker shock, and too worried about what else I can print and bind to recoup my recent purchases.

The one good benefit of staying up late night working on high end crafts is that I get to watch the likes of Monster house, Monster garage, and American Chopper.

But, that's another story.

 

Published Tuesday, April 20, 2004 1:51 AM by wadams

Comments

 

Robert W. McLaws said:

Here's the problem with the analogy though.The "built it" mentality is fine for hobbies and working in your back yard, but when it comes to your paycheck-earning job, that is exactly the wrong mentality to have. Especially if you are a consultant. Your time is precious. As a component vendor, I'd rather have you buy my stuff then write it yourself, and it's not just because I need to eat and pay the bills. It's because I want you to have time to go home to your family, to relax, or whatever. I spent 60+ hours working on a component so you could get a job done in 40 minutes, bill the client for 4 hours, and still make $90 on the deal and have time to make it to your kid's soccer game.

Now, why would most people rather build than buy at this point? I think it is because of a perceived lack of quality, and the pervasive nature of the developer ego. "I could build it better than these schmucks." The difference is, a master woodcrasftsman knows when to break out the table saw instead of using a handsaw. A good craftsman knows when to use the right tools. He wouldn't chose a hammer he bought over one he made purely out of ego.
April 20, 2004 2:12 AM
 

Shannon J Hager said:

In a perfect world, "buy" would always be the right decision. But far too often, the "buy" option means that you have to compromise beyond an acceptable range.
A B&N calendar is only $4, but it doesn't meet the first requirement: Yasmin's photos. Unacceptable, but the price is great. Kinkos has a better deal than $741, but it doesn't quite offer the level of control needed. However, the price-to-feature ratio would probably suffice, as would the calendars at cafepress/cafeshops.
Luckily, Yasmin's calendar doesn't have to sell enough to make the money back.
Pre-built programming components are great if and only if they do what you need them to do. A scrolling grid, for example, may have 24 features but if it lacks 1 that is required by the project, it won't suffice. Nor will a non-functioning spell check component. Or a free, built-in calendar control (with VS.Net) that produces horrible HTML that refuses to work within my pages.
I am pretty sure that I have tried more components that failed to meet my needs than components that have met them. After a while, you start thinking that spending 8 hours writing your own is better than buying a component and then wasting 4 hours getting the output "good enough", especially when there is a very real chance that the component will never be "good enough". If a component doesn't include source code, hacking something as simple as a background color for a textbox could end up making the component one of those so-called "negative income" generators.
Robert, how many hours did you waste on your spell-check component fiasco?
April 20, 2004 3:15 AM
 

Daz said:

As a consultant, i've been facing many choices between do and buy and the first thing that comes to my mind is ...
"How much work must i do to make this tool do what i need? Is it easy to connect with my existing applications?"

So here's a simple algo (probably too simple) :
If Tool.ChangeRequired > timeSpentToMakeItMyself
OrElse Tool.Maintenance = TooComplicated Then
Do it yourself!
Else
Buy!

April 20, 2004 5:28 AM
 

Joku said:

Shannon, very true. Many times components are built on the simplicity of use principle and lack ways to extend or modify functionality beyond what the component author had planned. Or like my last attempt to use a Zip component which ended up being horribly slow on old pkzip packed files cause of some difference in the format compared to the winzip/zlib etc files. First i tried #ZipLib, but it didn't even open those old Zips.
April 20, 2004 5:36 AM
 

Robert W. McLaws said:

Shannon- Way too many hours were wasted with that component. That's why I go back to my original statement that there needs to be more work done in the industry to provide a predictable experience across vendors.
April 20, 2004 1:33 PM
 

William Adams WebLog Build or Buy | Paid Surveys said:

May 28, 2009 6:56 PM
 

William Adams WebLog Build or Buy | Uniform Stores said:

June 1, 2009 9:36 AM
 

William Adams WebLog Build or Buy | Patio Chairs said:

June 1, 2009 3:02 PM
 

William Adams WebLog Build or Buy | Menopause Relief said:

June 8, 2009 8:02 PM
 

William Adams WebLog Build or Buy | low cost car insurance said:

June 17, 2009 12:35 AM
 

William Adams WebLog Build or Buy | Outdoor Decor said:

June 19, 2009 12:27 AM
Anonymous comments are disabled

© 2009 Microsoft Corporation. All rights reserved. Terms of Use  |  Trademarks  |  Privacy Statement
Microsoft
Page view tracker