Friday, October 27, 2006

Browser Stats

Now that FireFox 2.0 has been released, time for a another Browser Stats check to reveal what your typical non-tech is browsing with...

From Yesterday:

Internet Explorer: 86.64%
Firefox: 10.51%
Safari: 1.74%
Netscape: .47%

Internet Explorer Types:
6.0: 93.61%
7.0: 4.38%
5.5: .55%

Firefox: 73.68%
1.0.7: 8.27% 3.76%
2.0: 3.01%

Monday, October 23, 2006

What are the odds?

I'm not sure if they assign 800 numbers based on company name, but ever since I got an 888 number for SearTech around this time last year, I've had countless calls from people who are looking for Sears Technical support. Evidently the southern region of Sears hardware support used to have my 800 number, and someone there still has the number in the system to transfer people. So I get an unbelievable number of angry people who have been sent around their phone system... and of course they don't believe since I answer the phone as:

Me - "SearTech, this is Phil"
Angry Sears Customer - "Yeah, hi this is the FIFTH time my refrigerator has broken and I can't believe how many times you guys have transfered me around and..."
Me - "Actually, let me interupt, this is SearTech, not Sears Tech."
ASC - "That's who I'm looking for, because my #$%^%# Fridge is $##%5.."
Me - "No no, SEAR TECH, we make software..."
ASC - "Well, can you transfer me to the right number???"
Me - " sorry."
ASC - "$%$#%$!@"

What are the odds that SearTech gets Sears Tech's old number?

Friday, October 20, 2006

Browser Stats 10/20/2006

I feel my used cars site is a good cross section of the Internet browsing world at large. It gets a lot of hits, yet not all of them are tech people. Everyone searches for a used car at some point. Because of this sample, I thought it would be interesting to start posting browser usage stats to see how IE7 gains.

Looking at the whole month of September (Things below 1% not shown for the most part):
IE - 87.39%
Firefox - 9.31%
Safari - 1.73%
Netscape - .83%

Of IE:
6.0 - 96%
7.0 - 2.21%
5.5 - .71%
4.01 - .03%

Looking at ONLY yesterday, since that is when IE7 was released:
IE - 86.67%
Firefox - 10.82%
Safari - 1.2%
Netscape - .65%

Of IE:
6.0 - 95%
7.0 - 3.23%
5.5 - .58%

Thursday, October 19, 2006

Love Letters from Nigeria

My new favorite 419 scam letter:

I am interested in your car advert listed for sale,pls email me
back at your earliest conviniece your last asking price,the area for
the pick up immediately the payment has been made and lastly the reason
for offering it for sale.
I will wait to hear from you soonest for me to have bes confirmation concerning the car i have to send someone there to test and ship it for me.please i need better job from you,am sorry if am too demanded i have to coz have fallen into many trash before ,plz i dont want that this time arround.
Tony Finesse

Kind of typical boring scam letter until the last three sentences...

please i need better job from you

Wait... I thought you were buying a car from me?

am sorry if am too demanded i have to coz have fallen into many trash before ,plz i dont want that this time arround.

What? This poor gentlemen obviously has bad luck, or very bad balance, because he has fallen into trash many times. Poor little bugger.

Actually, I can't help him, but evidently eBay can:

Paul Graham's latest...

Paul Graham makes some great points in his latest essay, about the top 18 mistakes that kill startups. Hopefully it's not actually ordered in terms of importance, since the list starts out in typical "Paul Graham made it up" style, but ends with some really good points.

Let's cover them one by one.

1. Single Founder - This is just silly. He argues that a single founder means you couldn't get anyone to believe in your idea. How about this... I don't want to share revenues with a buddy just for the sake of having a partner? He's not completely off base though. No one is 100% well rounded, and having a partner that compliments your bad points can be a good thing. In fact, if you look at the successful partnerships, its usually "one business guy" and "one tech guy". But if you grab another tech guy, and you are techy too, I think the partnership is more likely to end in fighting then success. Either way, I believe you can succeed as a single founder.

2. Bad Location - I've covered this in previous posts, but again Paul reveals that he lives in the Paul Graham snow globe. Let's look at his ranking for top startup towns:

"Silicon Valley dominates, then Boston, then Seattle, Austin, Denver, and New York. After that there's not much"

Boston??!!! The number two?? Are you joking Paul? I know you live in Boston, but give me a break! In fact, outside of the Y combination companies, I can't think of a single startup from the Boston area. I can think of quite a few from Seattle and Austin though. New York seems like an odd pick too. I think in terms of startups per capita, New York is not a very big one. I know of a lot more startups in the DC area then Boston and New York combined. He uses Chicago as a counter example, yet I can still think of at least four tech startups from Chicago, and still zero from Boston.

Bad Location will not kill you. If you want Venture Captial, it helps to be in an urban area, but if you are a mISV, it really doesn't make a difference where you are, and I know quite a few successful startups in the boring suburbs of Maryland.

3. Marginal Niche - This is a good one. I have experience with this one. Chimney Sweeps are an extremely marginal niche. You will never retire selling entirely to Chimney Sweeps, I realized this from day one, but the plan was and still is to expand from Chimney Sweeps using the basic ChimSoft engine.

4. Derivative Idea - Paul...seriously. Where is your evidence here? Most of the BEST startups come from Derivative ideas. Google. Microsoft. Sun. Oracle. FogBugz. All of these companies made a better version of something that sucked. Better to improve an existing market then try to create your own.

5. Obstinacy - I probably would have had to look that word up...but yeah, don't be too stubborn to change your plan mid-course. Especially when it's your first company. I have NEVER gotten it right on the first try.

6. Hiring Bad Programmers - No argument or praise here... certainly this is more important in your early hires then your late hires.

7. Choosing the Wrong Platform - Ironic from someone who wrote a web based store in Lisp... ;) ... This is only a problem if you cannot handle your platform. Users have no idea what platform you are using. Reddit started in Lisp and MySpace started in ColdFusion, but neither of these mistakes KILLED the startups. If you decide to write a new web CRM in C++ instead of Ruby On Rails, ensure you can make the switch when you need to

8. Slowness in Launching - Here is where he starts to hit his stride. I've known quite a few developers in person that just NEVER get around to launching. Don't be afraid! Launching your product, even if you don't think it's complete gives you such motivation to improve it or at least move on.

9. Launching Too Early - No personal examples of this, but I've certainly seen company's rush a half assed product to the market only for it to be quickly forgotten (*cough* 80% of Google Labs *cough*).

10. Having No Specific User in Mind - So true. Usually when people come to me with brilliant business ideas, this is where I punch a hole in their plan. "Well, WHY exactly would someone pay our company $50 to place an add for a yard sale when CraigsList is free? Who are these people finding us?"

11. Raising Too Little Money - No comments here, I don't think this is KILLER, but it may make the difference between you being Google vs you being a much smaller company that still does alright.

12. Spending Too Much - Should be obvious, but some companies go nuts hiring people and building infrastructure while forgetting about the core business.

13. Raising Too Much Money and 14. Poor Investor Management - No comment here either, good points.

15. Sacrificing Users to (Supposed) Profit - Another mistake early startups make... pay attention to what a user wants. If you yourself would not buy or use your product, maybe you are not alone?

16. Not Wanting to Get Your Hands Dirty - I have this problem, and I think it's what kills so many startups. You must put yourself out there and be bold about getting sales contacts. Google AdWords won't work for every business, being social and selling are very important differentiators.

17. Fights Between Founders - Right...see my counter to point one. Don't start businesses with friends.

18. A Half-Hearted Effort - Pure gold here. I've seen so many people and startups give up if their product doesn't succeed in the first month. Things take time. Things take hard word. Occasionally companies get dugg on their first announcement and go public two months later, but occasionally people win the $200 million powerball. Don't play those odds. Figure out why your product isn't selling and improve it. Don't give up unless you have exhausted every possible avenue. Hard work is underrated these days.

Wednesday, October 18, 2006

Why I hate Peter Norton

Actually I don't hate Peter, but come on what programmer puts a picture of himself on the program's box? Really, my beef with Norton today is Norton's Internet Utilities. Customers started complaining that there was no link to view their ads. I didn't know what was going on, and when I got screenshots that physically showed the link missing I was really shocked. Even though the same browser was being used, a link was literally not there!

Finally after getting the HTML source from the customer, I saw some bizarre JavaScript on the page that wasn't mine. Well this explained it. Because the link to the ads was called "AdView.aspx" Norton blocked them. Argh. So I guess the moral is, beware using the word "ad" on your website. I guess I'll have to purchase Norton now just to test all my websites in now.

Sunday, October 15, 2006

If you want something done right...

The coolest thing about being a developer is if some application doesn't work the way you want it to, you can take matters into your own hands and fix it.

I had been using Yahoo! Autos maintenance for years now (maybe since 2000?). It used to be semi useful back in the day, and would figure out when you were due for an oil change and remind you. But then that stopped working, and they redid the interface, and ever since it's been horrendous to use. I saw they were hiring for that area a few years ago, so I assumed a web 2.0 makeover was on it's way, and I kept using it.

But still nothing has happened. After changing my oil this weekend I thought, jeez, this application is so simple, I am going to make something better myself. Since I have a car related website, I figured others who are as anal about car maintenance as me may want to use it too. So I spent a few hours this weekend throwing together a quick CRUD web app that already works better then Yahoo!. I still have a bunch of things I want to do before I release it to others, but in a matter of minutes I was able to get my auto history entered, and I'm quite pleased:

I hope to have it out by the end of next week, if you are interested in testing it, let me know. Security is always my biggest concern with apps like this... I spend probably 50% of my development effort thinking of scenarios where people might hack it.

Monday, October 09, 2006

New Home

A while back I was looking for a new place for my servers to live. Well I found a place, and today the move went off without a hitch.

The new hosting facility it much better then any previous ones I've been in, since it appears to be jointly ran by Lockheed Martin, and hosts government servers. After seeing the huge redundant generators and 3 story cooling towers with gravity fed water tanks I was pretty much sold. There is so much redundancy in this place, there is a good chance you will still be able to buy chimney sweep software and used cars if DC gets nuked. There's even 24/7 security in case you were thinking of physically unplugging my hardware.

My parting thought is that if I ever get my hands on the person that writes the interfaces for Netscreen/Juniper firewalls, I will rip them limb from limb.

Sunday, October 08, 2006

Have you had your chimney swept this year?

I get a kick out of telling people I make software for Chimney Sweeps. The first image that comes into most people's minds is Dick Van Dyke in Mary Poppins sitting there with a laptop. The truth is that Chimney Sweeps in this day and age are more like high tech inspectors then black lunged dredges from a Dickens novel.

Oh they still sweep chimneys, although it's usually done with a combination of brushes and vaccumes. Some even still dress up like the old time chimney sweeps. But they do a lot more then that. Today most chimney sweeps are equipped with chimney scanning cameras. These little robot-like devices can be sent up your chimney and reveal cracks and other fire dangers that would not be visible to most humans with a flashlight.

When a chimney sweep sweeps your chimney, they will typically give it an inspection as well, checking dampers, draft, and other fire dangers. Every year people experience chimney fires, and others die from carbon monoxide poisoning. Regular chimney sweeping can help to prevent both of these things.

With energy prices high, chimney sweeps can also help you insulate an area of your home that can leak lots of energy. They can install high tech flues, reflectors, and dampers to help you save more energy, and increase the amount of heat you receive from a fire.

I get my chimney swept once a year, have you had yours swept lately? If you live in the northeast I can probably recommend someone in your area, if not, be sure to check The National Chimney Sweep Guild's or the Chimney Safety Institute of America's websites for a certified sweep in your area.

Of course, if they are not using ChimSoft yet, feel free to give me a plug ;)

Thursday, October 05, 2006

Reward Yourself

For the other MicroISV'ers out there, if you are making $100 a month or $1k a month, remember why you probably started your business in the first place, to have fun! It's important to reward yourself in some way, or the stress will build and you will burn out quick. Set a small goal for yourself, and if you reach it, give yourself a small reward.

For me, I had always promised myself once I reached a certain level of earnings I would buy something I dreamed about as a kid, season tickets to by alma matter, University of Maryland Basketball. They are perrenially sold out, and season tickets go for a fortune on eBay. I finally got some this year, and not only will it force me away from the computer for 20 games, but it also reminds me why I am spending time on all this in the first place!