Tuesday, October 30, 2007

I've been a little busy lately

There is lots of good stuff happening at EnterpriseDB these days.  I did a webcast with Curt Monash (Monash Information Services) last week and it was a whopping success. 

Curt is a well-known database analyst and always tells it like it is.

We had over 700 registrations and almost 400 people on the webcast.  In addition to the 50+% of attendees that were decision makers from Fortune 500 companies, I was surprised to see a lot of state and local government and non-profit organizations on there.  I guess I shouldn't be surprised.  They stand to benefit the most from open source because they have less money to spend and need to spend wisely. 

The webcast was about choosing the right database for your enterprise application.  You can view the webcast on demand here.

I plan to pick up my blogging pace now and you can call me on it if I don't...

Monday, October 22, 2007

You are just a tool!!!

I have been thinking about the tools that make my job easier over the past weekend and those that certainly do not.  Here is a brief list of those that help improve my personal productivity.

  1. Trillian - Instant Messaging is an essential tool for me.  It makes quick conversations from one office to the next easier, and for those of us who tele-commute, it is almost like being there.  But, I have folks on Yahoo, AIM, MSN, etc.  Trillion handles them all very easily.  Nice tool and the basic version is free.
  2. Windows Live Writer - I am currently managing 2 blogs.  In addition to this one, I also have a blog entitled Obsessive Compulsive Reef Disorder, which talks about my reefkeeping hobby.  Window Live Writer makes it so easy to post to both blogs in a single tool with  point and click ease.  Adding pictures, inserting links are all a piece of cake.  The fact that it is free adds to its appeal.
  3. Microsoft Office - Being in an open source company, this one is always under debate.  In fact, we used Open Exchange briefly.  But, the rest of the known world uses Outlook and I can't live without it.  Its not the mail that is essential because there are many ways to handle mail, but it is the calendaring.  I let Office manage me.  If its not on my calendar, I will miss the meeting, plain and simple.  And scheduling a meeting is a piece of cake, too.  You can see when people are available and schedule accordingly.  Don't see a real replacement here yet.
  4. PDA/Phone - I have been searching for the ultimate combo unit for a long time.  I have had the Treo 600, Treo 650, Treo 700p and Treo 700w.  I loved my Treo, but was not a big fan of the Windows version.  It had promise but I was constantly forced to reboot.  And to make matters worse, we implemented Good Messaging at work for a while which disabled my standard calendar, email and contacts.  That really became a nightmare.  Now I have the Blackberry 8830.  This is my first crackberry and I can see the addiction.  It just works.  It does not have a touch screen.  But, I miss that less and less every day.  I am hoping that the next generation of iPhones will move the bar even further and be available on Verizon. 
  5. Netvibes - Netvibes is a great web 2.0 home page.  It has slowly replaced My Yahoo for me.  I can add multiple tabbed pages and get all the information I need in one spot from news, sports, weather and stock info to manage my eBay account, pull RSSS feeds and search videos and images.  I can even track this blog through it and analyze web stats about our corporate site.  Check it out.

My new favorite software by far, however, is Clearspace.  Its not a personal productivity tool, but rather a collaboration software suite for your intranet or extranet.  I am going to dedicate a whole post to Clearspace.  But, you can check it out here.

Thursday, October 11, 2007

Do Open Source Developers Fix Bugs Faster?

A recent article from CIO Magazine discusses the recent surveys done by Forrester and Evans Data Corporation.

According to the Forrester Report (sponsored by BMC), the average time to resolve an application problem is 6.9 days for enterprise developers and 6.7 days for software vendors.

The EDC report surveyed the open source communities and found that the average time between discovery and solution of a serious bug is under 8 hours for 36% of open source developers!  However, 57% of open source developers say that typical (not serious) bug fixes take more than 2 days.

I recommend reading the article by Esther Schindler.  She gets into all of the complications that enterprise developers have to deal with such as meetings, prioritization, meetings about meetings, etc.

It is a worthwhile read.

Wednesday, October 10, 2007

Can you make money from open source?

The question asked below was:  Can open source software be used to create a successful business model and generate real revenue?

And the full answer was:

I am assuming that you are asking the question from the point of view of building a business based on open source software, not just using it within your organization. Given that, I think the answer is a very large...it depends. And I would ask you if it matters in today's world.

For those that have been successful and profitable, look at the obvious ones, MySQL and Red Hat. Both very successful so far.
But, you also need to look at the ones that are not yet profitable that made out well via acquisition. XenSource was bought by Citrix for $500 million on $3 million in revenue. JBoss was purchased by Red Hat for $350 million and they were not yet profitable. Zimbra was just bought by Yahoo for $350 million on about $3 million in revenue.

The money is definitely there for sure. But, there are also a ton of open source start-ups who will never make it and the reasons are almost infinite. The two that seem to jump out at me are:

1. The open source project they are based on is too much of a niche product and has no viable market.

2. The team building the company does not understand the model and fails to execute.   As I said there are a million other reasons, but they are the two that jump out at me right away.

What I can say is that this is not like the internet boom of the late 90s. VCs are not throwing money at open source just because it is open source. They are looking for a solid business plan and a path towards profitability.

Another important question is how do you plan to make money? There are many models here.
1. Simply sell support for an existing open source product: Zmanda
2. Add value by bundling components to the open source project and sell support: Red Hat
3. Build your own open source product and sell it: MySQL
4. Take an open source product, add significant value and close source it: EnterpriseDB, IBM Websphere
5. Use open source components to lower your development costs and bundle it "invisibly" into your product: TiVo

The model matters just as much as anything else.

All my Answers are the BEST!

He says, tongue firmly planted in his cheek.  But, seriously folks, I have been doing this LinkedIn thing for a while and I find it to be an invaluable tool to find old friends and co-workers and to connect and network with others in my field.  A week or so ago, I decided to check out the "Questions" section and stumbled upon this question: 

Can open source software be used to create a successful business model and generate real revenue?

I found it to be right up my alley! It was asked by Angelo Varlotta, a PhD student from Purdue University.

Well, today I found out that he selected my answer as the "Best Answer".  Very cool.  He also sent me a nice email. 

So, I am 1 for 1 with question answering on LinkedIn.  I think I will retire now, undefeated!!!  lol

Tuesday, October 9, 2007

This Just IN: Postgres is FASTER than MySQL

Ok, for those you "in the know", you already know that this is true. But, it has been the conventional wisdom for a long time that MySQL is the faster database (at least for read-only environments).

Well, it seems that everyday there is more and more validation that MySQL is not a faster database. And here are some links:

Converted from MySQL to PostgreSQL - October 8, 2007

Postgres Publishes First Real Benchmark(SPEC jAPPServer) - July 9, 2007

Tweakers Database Test: Dual Intel Xeon 5160 - November, 13, 2006

We know that each environment is unique, of course. But, I think we can all safely say that Postgres can and does outperform MySQL.

So, what does MySQL bring to the table? There are comparisons all over the net. Pay attention to them, however, because many are outdated and compare older versions of both Postgres and MySQL. (This tends to hurt MySQL more).

A recent and fairly decent comparison was led by Greg Smith and the PostgreSQL community a month or so ago and can be found here.

But, for those of you without the time, let me help. Here are some benefits of using MySQL over Postgres:

  1. With MySQL you aren't limited by those newfangled technologies like Multi-Version Concurrency Control (MVCC).
  2. With MySQL, you are also not limited to those unbearable and strict rules of the calendar. You can easily add dates to February to make it conform to other months like February 31 (and January 32 for that matter).
  3. With MySQL you get CHOICES. You can get Foreign Keys (InnoDB) OR you can have Full Text Search (MyISAM) OR
  4. you can do clustering (NDB), but you have to choose only one because each of those lovely features is found in different storage engines.

I could go on and on, but I think you get the point. So, why is MySQL so successful? MARKETING. PostgreSQL is a true community with no company and thus no marketing engine behind it. All of the success of Postgres is through word of mouth or the companies around it like EnterpriseDB.

MySQL spends a lot of time and money marketing their offering. Its time for people to see through the marketing and go with the only REAL open source database, PostgreSQL.

Monday, October 8, 2007

Why Subscription pricing is better for you and your customers

In my last blog post, I talked about the hockey stick and I mentioned that changing the licensing model for software can help eliminate the hockey sticks in sales.

I want to talk a little more about the subscription model because I think it is the way all software should be purchased.

In a perpetual model, the customer pays a huge up front "license" fee and then a yearly maintenance fee, typically 20-25% of the license fee.   In a subscription model, the yearly charge is the same.  There is no up-front "balloon" payment.

Let's look at this first from the ISV standpoint and then from the customer side. 

What this means for an ISV:

  1. PLUS:  In a typical subscription model, there is a recurring revenue stream generated that can be depended on year after year and therefore the business can be planned better.  There are less peaks and valleys because of a missed deal.
  2. MINUS:  The growth rate is slower because ISVs do not get the initial spike of revenue from the license.
  3. PLUS:  You may sell more because the yearly subscription is much lower than the up-front license, so therefore most companies can justify the expense. 

What this means for a Customer:

  1. PLUS:  No capital expenditure.  This means the decision to buy is not left in the hands of the accountants and thus the software can be purchased faster.
  2. PLUS:  Dependable budgeting because the price does not change from year to year.  Need to add more licenses?  Simply pay a little more.  No need to delay because of high licensing costs.
  3. MINUS:  Depreciation is not really an option at this point.  So, if your accountants were counting on writing down that "major" software investment, no such luck.

It is taking companies time to come around to this model because it is not something they are used to.  The reality is that it creates a partnership between the ISV and their customer.  The ISV wants the customer to do well so they will continue to use the software and possibly need more.  The customer has not been raked over the coals and therefore feels better about the relationship with the ISV.

By the way, this is also very similar to the Software-as-a-Service model with only one exception.  In SaaS, the application is hosted. 

Monday, October 1, 2007

Hockey Sticks and the Software Business

Well, another quarter is "in the bag" and its time to take a breath and pause if only for a second...  

Now, get back to work and do it all over again. 

One of the most interesting phenomenon in the world of software is the dreaded hockey stick.  For those of you who are unfamiliar with this phenomenon, it is the seemingly unavoidable scenario that results in 98% of your sales coming in at the end of the quarter.  While this happens in many different industries, it is very prevalent in the software business.   Below is a chart that shows the hockey stick in action. 


Why does this happen?

  1. The incentives of the sales team encourages it.
  2. The fiscal quarter end of your company encourages it.
  3. The fiscal quarter end of your customer encourages it.

The problem is that, in many cases, all of the above are aligned.  This alignment leads to the hockey stick.  Re-aligning the three items above can help to minimize the hockey stick, but it will not make it go away.  Companies have tried for years and they have paid consulting firms millions to avoid it, but it is almost impossible.

What does the alignment do? Well, for one, your sales people will do everything they can to close deals before their clock resets.  That means they want to pull in any deal they can before the end of their measurement period.  Some companies have tried capping the comp plan, but that is not a good idea because then sales get pushed out causing a whole new range of issues. 

The second and third bullets are linked together.  Most companies follow a fiscal year that matches the calendar year.  They know that you do too.  So, they know that they are likely to get the best deals if they wait until the end of your quarter to sign.  They know that you will pull out all the stops to close that deal.  Combine that with their fiscal quarter and you end up with a HUGE hockey stick at year end.  Why?  Well,they have budgets that they have been holding on to.  They need to spend that money before year-end or they risk losing that budget next year. 

The combination of two and three are catastrophic to both businesses and personal lives.  Businesses can't plan and personal lives are ruined due to the fact that everyone has to work twice as hard from Christmas to New Years (when everyone ought to be off).  Not to mention the stress and the heart attacks that it causes. 

How do you fix this?  I am not sure you can.  But, you can mitigate its impact.

1.  Change your incentives to sales people.  Don't give them all the same quarter.  Make Tom's year end in May, Leo's in August, Mary's in March, etc.  This way they are closing deals all year.  Is this a new idea???  You tell me.  I have never seen it done, but no one could give me a good reason why not.

2.  Change your fiscal year-end.  My first employer did this (QAD) to lessen the December 31 hockey stick and it worked somewhat.

The best way, in reality, to avoid this is to change your business model. Part of the problem is the way software is licensed.  The traditional software model of a huge up-front license fee and recurring 20% maintenance helps to perpetuate the model. 

If there is one thing to be learned from open source and software-as-a-service (SaaS), its that a subscription model is the preferred way to sell software. But, that is a topic for another day.