Archive for November, 2009

28NovTechstars Update

Yesterday, I posted about some concerns I had about Techstars. The post came across a bit more harsh than I intended (I’m a coder not a writer), and I’ve received a lot of feedback from both the Techstars crew and several of their participants. However, some of my concerns still stand.

$6k per founder is not enough to be the sole source of seed money. Based on a few quick exchanges with David Cohen at Techstars, it’s clear that this is intended. The capital provided is to help, but not sustain a company. While I agree that Techstars shouldn’t be fully funding their companies for a long period of time, I think that covering the cost of living for the founders for the 3 month course wouldn’t be unreasonable.

On a separate note, part of my concerns are clearly a result of my misunderstanding about the purpose of seed money. Most of the events I’ve attended are focused on companies that have already started and are looking for Angel or VC funding. At these events, it’s always been implied that seed funding is generally enough to get the company running and ideally have a first product ready to ship. Based on what I’ve been reading, it looks like most organizations that provide seed funding are more focused on providing connections and expecting the founders to front the capital.

In the end, there are really 2 core problems:

  1. Organizations like Techstars work best for founders who already have some capital, rather than founders who are just starting.
  2. There is a lot of confusion and mis-information around about how start-ups get going before they receive Angel or VC funding.

Based on these problems, there could be a variety of solutions.

For companies like Techstars and Y-Combinator, it would be interesting if the system would still work if founders recieved $10k-$12k. This would cover cost of living for the founders for the 3 month program, but is still low enough to encourage them to keep moving forward. Based on the potential gains, this seems like a relatively inconsequential amount and it would open the doors to a lot of talented people who don’t have access to capital.

It would also be great to see a few events targeted at pre-seed entrepreneurs. It’s clear that Techstars and similar groups have worked for lots of companies. It would be interesting to have  day to hear from organizations like TechStars, CRV, Y-Combinator and some of their companies to hear how others have made it work. It would also be great to give people an  idea of what it really takes to get though the seed phase. There are a lot of people like me, sitting around with prototypes and plans with no idea how to get that first beta to market or prepare for Angel or VC funding.

I owe the people at Techstars an apology. While their program doesn’t seem like a good fit for me in my current situation, it’s clear that they’re doing a lot of things right. It’s just frustrating to see great ideas that could become multi-million dollar companies fail because the founders can’t borrow that first $30k. It’s tough to walk through BarCampBoston or Ignite and see dozens of great ideas that will never see the light of day.

in Rant. 3 Comments

27NovI Still Don’t Get Techstars

Update posted here

EDIT:  The tone of this post was clearly a bit off from my original intent. My main point was that it seems that the founders involved here are pretty much going all in by quitting their jobs for 3 months, whereas the investors are taking an incredibly small risk compared to other investment routes. Most of the other funding sources I’ve seen (Angel, VC, Bank Loans) at least provide some protection for the founders. (End of Edited Text)

As a developer whose participated in a few business plan / start-up competitions, it seems like everybody and their brother tells me to look into Techstars. I was pretty excited at first. Seed capital to get started and access to mentors, seems like a good idea right? After reading through the details, it’s definitely something I have no interest in. In all honesty, I think they’re misrepresenting themselves for a few reasons.

First of all, they prominently display $18,000 as the available seed amount in several locations. The amount is actually $6k per founder. Again, at first it seems like a great deal, but this is $6k/founder over 3 months. In the Greater Boston area, $2k per month is barely enough to cover rent and utilities and only slightly more than Wal-Mart pays. There is one difference though, Wal-Mart doesn’t ask for equity in your ideas. This leads to the second point.

Techstars requires 6% equity in your company. This seems a bit steep for a $6k investment. I know it’s very early seed money, but valuating the company at only $100k seems pretty lame given their rigorous entry process.  Since 60% of the companies have received Angel or VC funding at the end of the program, which was likely at least $1,000,000, Techstars must me pretty damn lucrative.

Keep in mind that the Manchester Young Professionals group managed to scrounge up $25k per business plus another $25k in services.  A small professional group in Manchester can front more than a group of investors in Cambridge?

I’m not expecting that Techstars should give money without gaining equity in the company. Trading equity for seed funding is common. However, this seems less like an organization that wants to help business get started and more like an organization that charges 6% equity to give you a meeting with some Angel and VC funding sources. Many of these investors are also likely at various start-up events that cost only a few hundred bucks. (See Speed Venture)

In the end, Techstars isn’t all bad; they’re just not representing themselves accurately. If you need $6k to get your company started, there are better places to get it. In the end, you’re not really getting seed money as much as you’re trading equity in your company for a few good networking opportunities.

I may be completely off-base here, and I’d be happy to hear from anybody at Techstars or their companies if I’m wrong.

in Rant. 5 Comments

23NovXenu Link Slueth

Man, I can’t recommend this thing enough.

Xenu Link Slueth is a great tool for checking a site for broken links and images. It also shows that a good tools can last, this thing has been out forever and I still find it to be one of my most used tools.

It rocks for several key reasons:

  1. It’s crazy fast compared to most of the web-based tools.
  2. The report it generates is incredibly detailed. It lists broken links by page, link and error code as well as providing a basic site map.
  3. It allows you to scan sites that are password protected.
  4. Above all else, it’s free.

Download it and show Tilman Hausherr some love.

in Tools. 1 Comment

20NovBlame SQL

We’ve all been frustrated by slow sites. We’ve all been to sites that would be great, if only they could render a page before I finish a game or two of solitaire. Many of us have also been on the other side; spending hours optimizing our sites just to cut the rendering time by a tenth of a second.

Developers, I’ve got some good news. Turn off YSlow and leave your javascript readable!  Stop debating between concatenating strings or using stringbuilders! BLAME SQL!

In many cases, I see developers ignoring SQL optimization. It may be because they’re more familiar with client-side tech or it could be the years of memorizing every design pattern they can, but almost every web application can benefit from a little TLC. Here are a few quick pointers.

1. Know what your SQL looks like.

Web developers rely on a lot of different tools and applications. Whether you’re using a large application like a CMS or a smaller ORM tool, it’s important to know what SQL your code is generating. In many cases, the SQL generated is optimized with only the most common use cases in mind or in some cases barely optimized at all. Don’t be afraid to peak at the stored procs your apps are using or do some logging to what your code is actually doing to your poor database. Just because you didn’t write it, doesn’t mean you shouldn’t know what it does.

2. Add some indexes

Having a basic understanding of indexes can help improve performance immensely.  To start off, make sure your tables have at least a Primary Key. Next, look at any columns you regularly use to filter selects or join on. Unless these column contain a lot of similar values, it’s a good idea to index them. Finally, if you frequently select a subset of data from a table, such as first and last name from a contacts table, you can create a covering index that will usually speed up these queries.

3. Look at your joins

 If you need a map to find your way through the maze of nasty joins in your queries, it’s time to optimize. First of all, make sure that you’re only joining to get data that you’re actually using. Joins can exponentially increase the time it takes to perform a query, so joining to get data that isn’t used wastes a lot. Also, when joining, it’s best to join on indexes. It’s also better join on numeric types than non-numeric types.

4. Subqueries are usually bad

This is a short one. If you’ve got subqueries, figure out how to make them joins. There are almost no cases when a subquery is more efficient than a join, and if you know those cases, you probably don’t need to take SQL advice from me.

5. Use stored procs for common queries

Stored procedures provide several performance improvements. Stored procs are partially compiled at creation and don’t need to generate a new execution plan at runtime. Additionally, combining statements into a stored proc can provide details to SQL Server that allow it to create a more efficient execution plan than passing individual queries. It may be tempting to just sling SQL at the database, but even a few stored procs can do wonders.

in Database. 1 Comment

18NovIE9 looks good, but bad developers are angry

Not long ago, IE6 was the dominant browser. CSS support was lacking, security was weak and performance was bad. However, with no competition, IE6 remained on top. Then Mozilla came along and things started to change. Mozilla and later Firefox proved 2 things.

  1. We should expect better than IE6.
  2. Microsoft can’t afford to have a crappy browser anymore.

So, with the help of the wonderful Molly Holzschlag and several other web standards gurus, Microsoft started to turn things around. IE7 was much less crappy than IE6, and IE8 is actually downright ok. However, with each improvement, the IE team seemed to get more flak rather than less.

Criticism focused in 4 areas:

  1. Poor CSS support (limited to CSS3, since IE8 has better CSS2.1 support than any other major browser)
  2. Slow Javascript performance
  3. Slow page rendering 
  4. Bad text aliasing

Today, the IE team released some early details on IE9.  Based on the early builds, the team has made massive headway in all 4 areas. Clearly the masses of developers that have nothing to do but complain about IE should be happy right?

Unfortunately, almost all of the comments are negative. They even include some great references to pngs (fixed in IE7) ,  Box Model (fixed in IE8) and IE6 (already addressed here). I would love to meet these “developers”. As a web developer, I’m expected to have a basic understanding of what browsers people are using and what features they support. However, it’s clear that these web developers are somehow still employed while being completely oblivious to 70% of the users on the web.

In the end, most of the people commenting on the IE blog are completely ignorant. The IE team is busting their ass to make a quality browser, and unlike most other browser vendors, they get shit on every time they make an improvement. I know that older versions of IE are a bitch, but crucifying the very people trying to fix things isn’t going to make things better and in the end may make things worse.

In closing, I want to thank everybody on the IE Team. You guys are working everyday to make my life as a web developer easier and you should know that many of us appreciate it.

To Dean, Ali, Bruce, Justin, Tony, Chris, Eric, Dave and the rest of the crew:

Thank you for you hard work and I look forward to IE9!

NOTE: Comments on this blog are moderated. If your comment doesn’t show up instantly, please refrain from sending me hate mail. I’m not blocking your comments unless they contain excessive profanity or derogatory statements.

in Browsers. 7 Comments

Blogroll

Recent Listening