Startup life in NYC. A pagerank 0 blog.
Tech Industry
Technical debt
01 year ago
In my last post I mentioned a term called "technical debt". It's common enough to have a Wikipedia article describing what it is. The article does a good job explaining "what" it is, but could use some clarification as to why it can be so nasty.
Technical debt doesn't show up in your accounting, but it can affect your bottom line just as heavily. Whenever you take a technical "shortcut," or accept a sub-optimal solution, you take on technical debt. In crops up obviously in the form of bugs, but more insidiously in the guise of design limitations. The larger a software system grows, the more internal "promises" have been made, that can't easily be changed. These promises are in the form of language choices, platform decisions, API's, 3rd party libraries, and the general structure of your system. A poorly designed system can easily find its way into a situation where it will legitimately take 6 months to implement a feature that only takes an upstart 3 weeks, working from a fresh codebase. That is where technical debt exacts it's interest, in development costs, bugs, and delayed launches.
On the other hand, it's called debt for a reason. It was taken out in a series of seemingly innocuous loans, but it can be repaid. Major platform refactoring or potentially a rewrite can be very costly, but are possible. During a major overhaul, almost no progress will be made from an external perspective. All of the work is going into fixing the internals. In most cases that I've seen, the best way to deal with this is to focus on designing new features in a better way, given what you've learned from past mistakes. As legacy systems cause problems, transition them over to the new design/platform, and slowly improve the overall health of your projects.
Please just don't ignore it. Nothing will kill developer morale faster than working in a codebase that they hate and are powerless to fix.
Why there will always be technology disruption
01 year ago
There will always be a hot new company in town, because technology equals disruption. Why? Because the latest technology is better, faster, stronger than the last generation. It's easier and cheaper to do more, and your cutting edge platform from 5 years ago is likely feeling its age.
Do all companies succumb to this problem? No, but building up and milking your flagship product for all its worth is tempting. Investing in the "next generation" of your own technology isn't always an easy pill to swallow when the current version "works just fine." The problem is that it may work fine for what you are currently doing, until a young upstart comes to the game with a desire to win and no existing customers to worry about pissing off or cannibalizing. They're riding on the last 5 years of general technology innovation and best-practices, and their going to eat your lunch technologically. You are now the incumbent, and they have a bigger gun and a terrifying hunger in their eyes.
The next 5 years will hold more technological progress than the last 10. Disruption is going to be coming faster and faster in every industry. The general "health" of your technology is going to become more and more important. Technology decisions are going to need to be made with regard for future development. In the software world, this is referred to as "technical debt."
Why Google can’t build social apps
01 year ago
Pretty interesting post I ran across on why the author thinks that Google can't build social apps. Some pretty good points. Google seems to make a point out of being utilitarian and productivity-friendly, but at a cost to stickiness.
http://ifindkarma.posterous.com/pandas-and-lobsters-why-google-cannot-build-s
July 2010 NYC Tech Meetup Review
01 year ago
I attended my first NYC Tech Meetup last night, and was really pleased with what I found. It was certainly a nerdy crowd, but much more diverse than I had guessed it would be. There was a certain hunger in the air. I don't imagine there was a single person in attendance without at least one big idea in their head that could "make it big." The company demos were overall pretty neat as well. Mostly hits, with a few misses in there. I haven't used any of these services myself, so the following impressions are based solely off of what I saw of their demos.
BetterFly helps to facilitate self-improve by connecting users with "betterists" aka personal coaches/instructors of many different expertises. Certainly an appealing way to plan and execute your self-improvement roadmap. Josh Schwadron did an excellent job presenting, and lead off the night with a polished demo.
Foursquare for activies. Seems like a bandwagon-app in which people share what they are doing with each other. It seems like a cool model, once they get adopters. Lots of monetization potential through biz dev with event organizers and content owners. The presenting team seemed to have a clear direction in mind, and had clearly done their homework on what is required to pull a service like this off.
Financial info, guidance, and tools, geared for women. I'm jealous. The website looked superbly crafted, and very well thought out. The presentation went really well, and I was left with the feeling of "Damn! What doesn't it do?!" Certainly a great way to get up to speed on personal finance fundamentals and get your money situation in check.
An open-source scripting and markup language to simplify the creation of RIA Flash sites. Bogged down by technical difficulties, and lackluster demos, the Frontal demo did not go as well as I'm sure the founders hoped it would. The technology proved inconsistent, failed to work several times, and in a grand finale, froze a Mac so bad that "Force Quit"s and a reboot were required in order to continue with the rest of the demos. The software seems interesting, but not quite ready for production use. In the meantime for RIA dev, I'm sticking with OpenLaszlo, Flex or GWT.
Visual food check-ins. Want a certain type of food? Check out pictures of the eats from nearby restaurants before you go. The demo was quick, and all in Powerpoint, so it's hard to really get a feel for the app itself. I'm on android myself, so this one's getting the pass from me for now. Maybe once they come out with an Android version I'll give it a checkout.
If I was single, I would have signed up for an account within 12 hours of the presentation. The site is an online dating service that allows people to propose dates, and meet up with other single people that are interested in that date as well. It is a terrific idea, and seemed to be phenomenally executed. It was quite apparent that the founders are "not just the founders, they're also members" and have been having great success with it. If you're single, or just looking for great date ideas, I'd seriously reccomend checking it out.
Their basic premise is to implant a flash widget everywhere that allows people to purchase products, inline on the site. I think. The whole presentation was really disjointed, and lacked focus. I feel that we were told "This is the future of eCommerce!!!" at least 3 times, but I don't think anybody bought it. Just because it is right there doesn't mean that it's where I want to purchase. It feels like a sketchy dude in a trench coat, trying to hustle watches outside the front entrance to a mall. No thanks, I'm already at the mall. I'm going inside. Just because it's right there doesn't mean I'm biting.
Beautiful, clean interface for luxury travel inspiration and booking. Built upon (and incubated from) the Gilt shopping model, JetSetter seems like a solid site that focuses on keeping things simple, useful, and narrowly directed at their target audience.
I don't even read comics and I'm inclined to check out this product. Phenomenal technical execution, hard-hitting business integration, and a slick demo/presentation. I hear that these guys are doing quite well for themselves, and that doesn't surprise me in the least. I have a feeling that these guys are going to keep the momentum strong.
For $10, the event was a great experience. Lots of useful tech insight, and plenty of networking opportunities. I hope to be back to plenty more in the future.
Google is going to engulf the internet…
31 year ago
…like ants over a carcass.
Google's claim it fame has been it's ad platform historically. What a lot of people haven't talked about, however, is their architecture as a platform. Google Web Toolkit and the Google App Engine are creeping their way into websites. Google search has infiltrated thoroughly the consumer population. What they have left to do is control the business IT side as well.
I'd put 10 to 1 money on Google having plans to release their own web server to the public. Their strength is their architecture and knowledge of web standards. Imagine every website running on Google software? They would understand everything about the internet. They would literally be able to watch everybody walk across the internet, with their own unique VIN. They would be able to tell down to the person what ads to display. They would know as much about you as Facebook, without you having to put that info in.
Google could be on the path to becoming big brother. Please don't be evil.
Edit: Kurt K. let me know that Google does indeed have their own OSS webserver in the wild. I'll be very interested to see where they take it.
Why Google Chrome will toe-off with the “Like” button
02 years ago
There’s been a lot of blog buzz about what Facebook’s “Like” button is doing to the web. A good deal of people have even been predicting the beginning of the semantic web. I’m not going that far yet, but Facebook’s attempt to ”Build[…] a web where the default is social” is a pretty lofty aim. They’ve already had a ton of success getting 3rd party websites onboard with their like button, and they seem poised to step in as the default “identity platform” for the web. One sign-on, one profile, all through Facebook.
All roads lead to Rome?
This is great news for Facebook if they pull it off. Unfortunately, the web as a whole isn’t going to come along quietly on this one. All traffic coming from and going to one source? Information is too pervasive, and the web is too democratic. Websites will put up the “like” button to enjoy the nice flow of traffic, but it will be a long time before Facebook convinces everybody to suck off their teat for all of their traffic.
In the meantime, Google has a major opportunity. The strength of Google is in their tendrils across the web. Facebook is a Juggernaut that is starting it’s advance across the web, but Google is already there, and needs to dig in deep to defend its claim. The “Like” button requires a lot of content providers to buy into Facebook’s Open Graph. Although sites are signing up in droves, it will be a long time before it reaches critical mass. Not everybody wants to commit allegiance to Facebook, and not everybody wants to be a part of the “social web.” Google has the tool to beat Facebook to the punch in Chrome.
A truly open graph
Imagine a “Like” button on Google Chrome. Simple, pervasive, and works on every site. The difference is that it would work with any sharing platform. Facebook, Twitter, Buzz, your blog, whatever. Facebook can build it’s Open Graph with Facebook in the center, but Google has the capability to build a social graph of the entire web, without Facebook’s geocentric skew. It's wildly simple, but incredibly powerful. Google can get into the graph game strongly by supporting every platform out there. They can stay on the cutting edge of distribution channels, and can keep weaning people into the Google fold.
Google isn't about search. Google is about distilling large amounts of information into a positive user experience. Having access to an web-wide social graph empowers them to take the internet one step closer to semantic. To be honest, I would be amazed if anything I've talked about in this post isn't already in the works at Google right now.