Brad Wardell's Blog


Partners vs. lone founders

Published on Tuesday, August 2, 2005 By Brad Wardell In Business

A long-time friend of mine is facing real difficulties right now over at the site he helped create -- deviantART.com.

It's almost a text-book example of business men vs. engineers on the surface.  The idea guys, the founders, the visionaries versus the suits, money men, and biz guys.

In these situations, the biz guys almost always win because they are the ones who have put together all the legal paperwork for the company and can make sure they are legally unassailable by founders who, too often, agree via a handshake or via email.

At Stardock, I founded the company alone. I was both the engineer and the biz guy.  I don't have to contend with a board of directors or shareholder votes.  This comes up on occasion when some infuriated customer or user sends a nastygram to our customer service department because I've told them off or whatever.  Typically, companies don't reach the size of Stardock without investors and business developers.  I would be the first to admit that we'd probably be a lot more successful financially today if we had a few suits helping to run things.  But next time I'm tempted, I'll remind myself of what Jark has gone through.

We now enter the realm of raw speculation.  But back in 2000, a couple of guys had an idea -- to create an art community called deviantART.  It would be a bit different than other sites in that the focus of the site would be about the artists and not the users. This was a radical idea at the time.  When Stardock created WinCustomize.com the focus of the site was on the users first and the artists second.  It was assumed, with I think good reason at the time, that there's a lot more users than artists.  But deviantART believed that there were millions and millions of would-be artists out there and no central place for them to congregate.

As a result, deviantART automatically gave each artist their own sub-domain and encouraged them to interact with one another.  For instance, mine is frogboy2.deviantart.com.  The primary visionary in all this, from what I recall, was a 28 year old visionary named Scott Jarkoff.  With his friend "Matteo" they coded the basics of the site. 

The site grew at a steady rate until they teamed up with a very young guy named Angelo who was, I believe, around 19 years old at the time.  From what I've seen, he is a gifted young business man.  And he was able to provide the site what amounted to unlimited bandwith.  He also brought in guys who created a network infrastructure to support much higher traffic.  The site was relaunched and quickly grew from being an Alexa ranked site of around 6,000 to an Alexa ranked site of 800 (800th most popular in the world).  Today it's in the top 500.

Both guys deserve immense credit for what they've done.  But there is always the question of ultimate control.  At the end of the day, one guy is going to call the shots at any company.  And that guy became Angelo (Spyed).   Over time, he came to control not just the behind the scenes business but slowly emerged as the front-man for the company.  As a result, he tended, especially in the media, to get the lion's share of credit for deviantART even though it was really mostly Jark who had the vision and overall idea and most importantly, the elbow greese to put the site in a position to go to the next level.  Jark and Matteo created deviantART.  Angelo took it to the next level. 

A lot of the issue today has to do with a basic disagreement -- what was Angelo's involvement at the beginning? Can he really be considered a co-founder?  That's not for me to say.  But I don't think it's accurate to put Angelo on the same level as Jark and Matteo.   To me, the issue is simple -- remove Jark and would there have been a deviantART? The answer, I believe, is no.  Take Angelo out and there's still a deviantART albeit a much smaller, niche site.

Stardock today is a multi-million dollar company.  But if some big time business guy came on and took over as CEO and took Stardock to being a multi-billion dollar company, would it be right to call him the founder?  After all, if Stardock were a $10 billion company, then of how much significance are the days when the company's grosses were measured in the low millions range? Is that time really significant next to the billion-dollar verison? 

How you answer that is probably similar to how one would look at the deviantART situation.  deviantART today brings in around 20 million unique visitors per month.  Once Angelo took over, the site went from being a 300k uniques per month to a 3 million uniques per month within a year. How significant is the time when the site was really little more than a niche art site with some cool ideas? 

I'm an engineer so I take things literally. Jark and Matteo founded deviantART and Angelo made it a business success.  Jark's motivation was probalby doing something unique, cool, and interesting and his initial desired reward comes in the form of kudos, acclaimations, and glory (for lack of a better word).  Hence, I suspect that that what stings Jark more than the money is that Spyed (Angelo) seems to have rewritten things such that he was the visionary and driving force behind the site rather than Jark.  As a result, Angelo's in the position to reap both the financial benefits and the glory of the site.

Don't underestimate glory.  I confess that I myself worry more about "the community" than making "big bucks" very often.  Money is merely a score-card, recognition matters more to me and I suspect Jark is much the same.  Perhaps from Jark's standpoint, Spyed is attempting to get money and recognition for things that he hasn't warranted said recognition.

I think we'll know more over the coming weeks.  In the meantime, I guess I'll take solace in the words of my wife "See, it may a bit of a bummer not to have partners to have helped build the business but you don't have to worry about someone usurping me."  Well, not for a high price-tag anyway.

 

 

Natural Desktop 2006

Published on Tuesday, August 2, 2005 By Brad Wardell In WinCustomize News

Stardock has released Natural Desktop 2006 v1.5 today.  Natural Desktop is a desktop enhancement program that looks at your current local weather, the time of day, the time of year and then applies a wallpaper, sound track, and animations based on that data.

If you're stuck inside working on a summer day, Natural Desktop is just the cure for that.  Listen to morning doves singing in the morning or cicadas in the afternoon, crickets in the evening. The sound tracks were put together by famed composer, Mason Fischer to sound just like what one would expect if they were outside.  The wallpapers were created by Ryan Bliss, owner of the well known wallpaper site, Digital Blasphemy.  The unique animated weather elements were designed and programmed by Matt Crocket, well known here as "Tiggz".

The program is immensely configurable with users turning on and off various elements of Natural Desktop along with being able to add their own sound tracks, wallpapers, and other parts to the included library of sounds and graphics.  Version 1.5 is a nearly complete rewrite by Laurence Parry who has gone through and implemented a host of optimizations and new features.  Natural Desktop is only $9.95. Proceeds from it help pay for WinCustomize.com and support other unique, cool projects of this kind.

New features & changes include:

  • Fast zip-code lookup that gives the user the option to find multiple local weather stations to choose from.
  • Will display an extended 5-day forecast for most places in the US.
  • Enhanced support for European and other international locales for weather reporting
  • Links to infrared and satellite images now available.
  • Clouds move at a speed based on the local weather conditions (wind speed)
  • Preferences now stored in local directory rather than registry for faster performance.
  • Lots of fixes, tweaks, etc.   

Screenshots:
  

Website:
www.naturaldesktop.com   

Windows Vista road-map

Published on Monday, August 1, 2005 By Brad Wardell In WinCustomize News

Head of all things Windows at Microsoft, Jim Allchin provides a heads up on the operating system formerly known as Longhorn.  The article at ZDNET is a bit..ahem..spin heavy with such amazing statements such as:

In the four-year history of Longhorn it's arguably the first time Microsoft can claim to be ahead of schedule. Since first discussing Longhorn in 2001, Microsoft has changed the features of the release and pushed out its launch date several times. But there is little room for error now, as Microsoft is trying to have a final version ready by next year's Christmas buying season.

But it's a terrific primer on when and how Windows Vista will come out.  Our own internal evaluation is that Windows Vista is going to be excellent.  It'll have all the strengths of Windows XP but have a new, more solid networking foundation, a much better presentation system, and a completely modernized API set for creating modern applications much more quickly.

Does JoeUser.com seem faster now?

Published on Wednesday, July 27, 2005 By Brad Wardell In Blogging

Does JoeUser.com seem like it's faster? In the past week in the course of trying to fix the site, we (and by we I really mean T-Man and Alberto) have been working like madmen to optimize the site code to make it more efficient.  Hopefully, getting to articles and such feels a lot faster.  How does it feel to you? It should feel significantly faster so if it doesn't, let us know. Thanks!

Windows Vista Beta 1 Overview

Published on Wednesday, July 27, 2005 By Brad Wardell In Windows Vista

The first beta of Windows Vista has arrived.  Windows Vista is the next generation of Microsoft Windows.  And Microsoft has, in many respects, gone back to the drawing board.  It's often forgotten that Windows XP came very soon after Windows 2000.  After the Windows ME debacle, Microsoft quickly concluded that they needed a consumer version of Windows that was based on the "NT Kernel".  The result was Windows 2000 with a few enhancements called Windows XP. 

Much has happened since October 2001 when Windows XP came out.  Worms, Spyware, Malware, and the like have become household names.  People yell at Microsoft for the lack of security in Windows without realizing that the insecurity of Windows wasn't born out of technical insufficiency but rather an excessively optimistic view of human beings.

Back when Windows NT was originally developed (early 90s), there was a sense that we were all one big happy family.  NT's main competitor, OS/2, was even less secure in many respects.  IBM's vision at the time included users sharing parts of programs together seamlessly via OpenDoc.  OS/2 included a powerful batch language called REXX that could have been used to create things that would make today's Worms look like child's play.  Back then, developers imagined users working together in harmony.  Kumbaya should have been the code-name of Windows NT.

Few people back then could have predicted the impact that over-wedgified 17 year olds spreading their angst across the net in the form of Spyware, Worms, DDOS attacks, and so forth.  But that was the world Windows XP found itself in.  And Windows XP was based on that Kumbaya NT kernel.  Fixing these security issues is non-trivial in most cases.  There's compatibility to take into account for one thing.  And so over the past 4 years Microsoft has gone back and rethought how software should be written and how networking could be done more securely. 

At some points, it looked like Microsoft might go overboard.  "Managed Code" became a rallying cry at some parts of Microsoft.  Managed Code is a nice way of describing taking away the developer's ability to actually create things for the OS that the OS vendor didn't already think of.  Luckily, managed code didn't become the order of the day.  Windows Vista, from our initial glances, looks to be a good balance between the need for security for users and the need for flexibility for software developers.  WinFX, the managed-code API is there, but it isn't required (at least not yet, and hopefully MS will always allow developers to bypass managed code to get to the lower level APIs).

The main highlights of Windows Vista can be summarized with:

  • Better Security.  This comes in the form of hardened services.  Essentially, the kinds of things that hackers took advantage of (buffer overruns and the such) should be a thing of the past.  The ability for malicious programs to patch system files and the such is being curtailed.  Moreover, Microsoft is trying to migrate users away from setting users up as administrators.  Instead, users set up with normal access can be prompted to enter the admin password when they need to do something that can affect system integrity. Hence, even if something does get through Vista's security, it won't be able to go crazy if the user isn't logged on as an admin.
  • Better User Experience. The code-name for the new user experience is Aero.  It's designed to be fast, efficient, responsive, and intuitive.  In that, I think they've succeeded.  I do suspect that the glass effect will need more tweaking before the final release.  Perhaps coloring the glass for active and inactive windows might help as well as decreasing transparency of the active window.  More strategically located shadows might also provide more clarity without compromising that improved aesthetics.  One correction, the user experience is not based on Avalon. It is its own thing.
  • The New Development Platform.  Microsoft has also created a new development platform called code-named Avalon (but has now given it a long boring official name).  Using .NET based languages (such as C#) along with a new mark-up language called XAML, developers can now much more easily (read more quickly) develop applications that have their own unique look and feel.  XAML has been designed to let developers design their UI in much the same way that a web developer makes a web page.   For developers, think of XAML as being similar to a .RC file.  You still have the .cpp file (or whatever language you're using) but .xaml is what defines the UI (one wonders whether Yahoo was aware of Avalon before their recent buying spree but I digress).
  • Information Organization. WinFS (the database layer for the Windows file system) won't be in Vista.  But Microsoft is working to make fuller use of NTFS's capabilities to make it much easier to find your things and organize your data than what was previously possible.  Simply put, searching for yourself will be much easier and faster.  The virtual folders feature looks to be very promising.  For those people who have been using short-cuts in painful ways, virtual folders may be just what you're looking for.
  • Windows Communications Foundation (WCF). Code-named Indigo, Windows Vista is designed to make it far far easier for developers to create programs that can talk to each other over the Internet securely and easily.  Developers (particularly game developers) can tell you the pain it is to write things that can easily talk to each other over the net through routers, fire-walls, etc.  Opening ports and such is not an ideal solution. Indigo, amongst other things, aims to solve these kinds of problems.  Over the past few years, a gazillion different protocols have come up that developers have had to juggle.  WCF lets developers focus on writing good software instead of trying to figure out how to talk to other computers.

From what little I've played with Windows Vista so far, it's quite snappy.  After I've had a chance to play with it more, I'll post some more observations.  The real meat, according to Microsoft, won't show up until BETA 2..

Related:
Windows Vista Pre-Beta preview
Stardock & Windows Vista

What the heck is the deal with JoeUser?

Published on Wednesday, July 27, 2005 By Brad Wardell In Blogging

JoeUser has been going up and down and up and down.  What's the deal?

This is going to be a bit technical but we want you guys to know exactly what is happening.

All Stardock community sites share the same underlying Stardock.net database structure.  Your UserID and Password will work on JU, GalCiv2.com, TotalGaming.net, WinCustomize.com, SocietyGame.com, PoliticalMachine.com and even in Stardock.com's forums.  The same is true of the blogs/posts.

When I write a blog here, it goes into a single master database which then gets replicated out to other databases (to increase speed).  Based on the category you posted, that blog will show up (get syndicated) to other Stardock.net based sites.

For example, if I write a blog on Windows Vista in the category Personal Computing, it will also show up on TotalGaming.net and WinCustomize.com.  This has been part of our goal to make sure people's blogs get read by people.  That was one of th eincentives to creating JoeUser.com in the first place -- most blogs never get read.  But here on JU, we make sure your words get out there so that you can interact with other people.

The problem, however, is that JoeUser.com was the first of these sites to be created with this system.  Or put another way, when we wrote JoeUser.com, we were just learning how to do this stuff. 

As the network of sites has grown in popularity and traffic, JoeUser.com has taken a bigger and bigger bite out of the overall resources available to ALL the sites.  Resources like CPU usages and database bandwidth.  JoeUser.com, which gets only around 5,000 unique visitors per day (which is still quite a bit) uses up over 50% of our ENTIRE network's resources in CPU and database resources.  The other half is shared by WinCustomize.com, TotalGaming.net, Stardock Forums, GalCiv2.com, etc. which get over 150,000 unique visitors per day (30 times). 

To put it another way, by itself, JoeUser.com was slowly bringing the whole network to a crawl.  And it still is.  Since we don't want to have to rush rewriting the site using the newer technology developed for WinCustomize and TotalGaming.net, we're trying to figure out preciesely what it is about JoeUser.com that does this.  And unfortunately, that means keeping JoeUser.com down when that happens.

We are getting closer to figuring out the problem but we can totally relate that it's very frustrating that we can't get to our blogs.  Hopefully within a few more days we'll be back and the site will be faster than it was before.

Thanks for your understanding and patience!