Is your head in the Clouds?

Cloud computing is all the buzz these days.  Amazon, who'd a thunk it, is leading the charge with their Elastic Compute Cloud(EC2) and Simple Storage Service (S3).

So, what is all the buzz about?  Cloud computing is the next generation of the ASP or hosted model.  It is not unlike a SaaS (Software-as-a-Service) model either. 

Essentially, you can load up your application to someone else's server and pay them for storage and bandwidth and utilization.  Sounds kinda like a web host, eh?  That's a good analogy.

Today, EnterpriseDB announced that we will be offering EnterpriseDB Advanced Server Cloud Edition.  Though it is currently a BETA program, the potential is huge.  Load up Advanced Server in the cloud and scale out/up as necessary without having to worry about maintaining the infrastructure and, even better, pay for usage only in small increments of time.  Instead of having to purchase a large server that is mostly overkill, but sufficient for the peaks in your business, you can run it all on the cloud without a worry.

Sounds like the future is just around the corner!!! 

IT Innovation in a Recession

With all the talk of a recession going around these days, I can't help but think about the post Y2K days when the bottom fell out of the tech market.  At the time, I ran a fairly successful ERP consulting practice.  My clients included:  Colgate Palmolive, Avery Dennison, Johnson and Johnson, Guidant, Arrow International, etc.

Business was booming through the first part of 2000.  It was a windfall of cash for my company and luckily, I squirreled some of it away for a rainy day.  I just knew it wouldn't last.  Why?  Well, the Y2K scare was over.  All of the software updates were done.  And, to make matters worse, all of the hardware purchases were complete.  Many of the applications that had issues were written in COBOL and, yikes, FORTRAN.  These machines were old and outdated and entire application suites had to be rewritten and upgraded to new operating systems and hardware.  So, the software was fixed and new hardware was purchased (at considerable expense to major companies.)

At this point, IT budgets froze.  They had spent their money and wanted to wait for some of it to pay off.  Then, the bubble burst on the dot coms and IT budgets got cut.

I remember having multiple 6-figure deals just going on hold until further notice.  They finally started picking up again in late 2002, but the damage to the tech industry was pervasive.

Now, here we are in 2008 and a recession is looming.  (I am not buying it, for the record.)  IT budgets are in jeopardy again.  When IT budgets get cut, innovation ceases and preservation is the name of the game.

But, that doesn't have to be the case anymore.  Evaluate your current IT budget.  Look at the planned expenditures, especially the largest software components.  Are they open source or proprietary?  If they are proprietary they are most likely capital expenditures and more likely to get cut.  If there was ever a case for open source, a recession makes that case even more clear.

Open Source software components are "good enough" if not better than proprietary solutions in almost every case.  Yes, your CTO knows and trusts X company (in our case, Oracle).  But, your gonna cut that whole project anyway because (Oracle) is too expensive in a recession.  Teach him about Postgres and EnterpriseDB Advanced Server.  Let him know how SugarCRM, Alfresco, Jasper, JBoss, etc.  Make them understand that they have been running and depending on the LAMP stack for years and that Firefox is actually open source.

If they don't trust you, get OpenLogic in there or SpikeSource or Optaros.  The cost of the open source software, plus the piece of mind of a third party like I mentioned above, will still be dwarfed in comparison to the cost of the proprietary solution you are replacing.

You CAN innovate in a recession and it might just be the competitive advantage you need....

Bill Gates at CES 2008, WSYP, Microsoft Marketing

I know this is a little old now, but in case you haven't seen it, Bill Gates gave his final keynote at CES a couple weeks ago.  For those you who have attended before, you know that his keynotes have been hit or miss.   The highlights included the launch of the XBox while the low-lights included a few BSOD (Blue Screens of Death) and who could forget Microsoft BOB.

This year, though, he may have sealed his legacy, not as a technology pioneer, but as a comedian.  This video is hilarious.  (Not hysterical.  Ask Paul Corson from QAD.  He was an English major and that was one of his pet peeves.)

The video looks forward to his last day at Microsoft and what it will be like.  It includes cameos by a whole bunch of folks including:  Brian Williams, Matthew McConaughey, Hillary Clinton, Jay-Z, and others.  Enjoy!

 

And while you are at it, check out these two videos that need to be resurrected and passed around.  This first one shows you just how marketing can eff up a product!!!

 

And this one shows you that Microsoft really does have a sense of humor...  The Microsoft WSYP Program (We Share Your Pain)

For those of you wondering what is up, I learned how to embed video!!!   And I have been thinking a lot about viral marketing.  These are all great examples of viral marketing!!!  More on that in the future.

Postgres is not for sale

Rather than try to summarize Greg Sabino Mullane's excellent blog post from yesterday, I am going to give you a link to the original post and copy the post right here.  Greg really nails it on the head.  He provides his commentary on the MySQL deal itself further down in the post, which I did not copy.  Check it out.

In light of the recent MySQL sellout, I'd like to once again answer the question that pops up occasionally: "Who will purchase Postgres?" Sure, it brings a smile to those of us immersed in the open source world, but it bears a serious answer: Postgres cannot be bought. While from a distance, MySQL and PostgreSQL look the same ("open-source databases"), they are very different beasts, both in technical and non-technical ways. In a nutshell, the difference can be expressed as:

MySQL is an open-source PRODUCT.

Postgres is an open-source PROJECT.

Only two letters of difference between 'product' and 'project', but a very important distinction. MySQL is run by a for-profit corporation (MySQL AB), which owns the code, employs practically all the developers, dictates what direction the software goes in, and has the right to change (indeed, has changed) the licensing terms for use of the software (and documentation).

By contrast, Postgres is not owned by any one company, is controlled by the community, has no licensing issues, and has its main developers spread across a wide spectrum of companies, both public and private. Can a software product succeed using such a system? Well, the other letters in the original LAMP (Linux, Apache, and Perl) have similar models, and they seem to be doing just fine. Like Postgres, there is no way to "buy" any of those projects either.

Of course, EnterpriseDB is a company.  And, we have an excellent database built on the Postgres foundation.  And, I may get in trouble for this, but I am willing to entertain all offers for EnterpriseDB starting at $5 billion US.

As noted at the bottom of the blog:  The comments expressed by me in this blog are my own and may not represent the opinions of EnterpriseDB.

Some wrap-up on the MySQL deal

Now that the dust has settled on the MySQL deal a little, I thought I would add a bit of my perspective.

First off, what is Sun going to do with MySQL now that they have it?  Do you think the $40,000 all-you-can-drink pricing will go away?  You better believe it. (Note to MySQL users, expect a price increase.) With that top-line price, MySQL had capped their own market.  Assume for a second that every Fortune 2500 company bought the enterprise-wide license.  That would mean 2500 companies to support and the annual revenue would have been $100,000,000.  That is not a lot of money when you consider that the database market is $15 billion and Oracle owns $8 billion and Sun paid $1 billion for MySQL. 

I don't think you spend $1 billion to make $100 million.  But, that has been Sun's way. 

Sun does not have a very good track record with acquisitions.  Here is a laundry list.  How many of those have been successful?  So, will Sun be able to make this happen?  I doubt it. 

And, what are they going to do about the storage engine nightmare that MySQL has found themselves in? 

  • MyISAM is read-only. 
  • InnoDB is owned by Oracle.
  • SolidDB was just purchased by IBM.
  • Falcon is still in alpha and YEARS away from being mature.

Even with an influx of cash and developers, the Falcon storage engine will need time to mature.  Cash and code can't change that.

Maybe these are the reasons that Marten went looking for a suitor.  The business model does not support a long term IPO success and the underlying technology is not owned by MySQL.  Compound that with the fact that for the millions and millions of downloads that MySQL flaunts, their ratio to paying customers is a number far lower than .1%.  And yes, the decimal place is in the right spot.  If it isn't, it probably should be .01%,  1 one-hundredth of a percent.

Seems to me that Marten made a great deal and Jonathon had his blinders on.  What could Sun have done if they invested half of that money in Postgres?  Not to worry though, EnterpriseDB is here to save the day...  ;)

PostgreSQL beats Oracle and MySQL for Best Database of the Year Award

Developer.com just published their 2008 Product of the Year awards and the audience voted.  For Best Database Tool or Add-in, the nominees were:  Hibernate, JavaDB, MySQL, Oracle Database 10g, and PostgreSQL.devpoy_08sm

Here is the quote from Developer.com:

And the winner is PostgreSQL. This one surprised us and taught us something about our audience base. We need to do a better job of covering this topic! PostgreSQL.org touts this product as "The world's most advanced open source database" and after seeing the votes, they may just be correct.

Congrats to MySQL

I know I spend a lot of time on my blog talking smack about Oracle and MySQL.  And, honestly Oracle deserves the smackdown far more than MySQL does.  I have met Marten a few times at shows and I have to say from my limited interactions, he is a really nice guy. 

And so, I would like to congratulate all the folks at MySQL for a job well done.  Let's be honest.  The goal of being in business is to make money.  And, the folks at MySQL have worked hard for a long time. 

Enjoy it.

What makes MySQL so popular?

In a previous post, I discussed the claims of both MySQL (The world's most popular open source database) and PostgreSQL (The world's most advanced open source database).  I said I would dig deeper into WHY MySQL is so popular. 

The answer is quite simple, really.  It is pervasive.  But, to fully understand how they got this way and where they are pervasive, we need to go back more than a few years.

MySQL is roughly 10 years old.  That means it was started in the late 90s.  This was right about the time that everyone and their brother decided that they could have a website.  For a brief history of the web, check out the Hobbs Internet Timeline.  As you can see, the birth of MySQL ties nicely into the explosion of the internet.  For example, 1998, the 2 millionth domain name is registered.  By 2000, there are over 1 billion index-able pages.

So, back to my story.  The web is exploding.  Thousands of static websites are popping up every day.  And, small web hosting companies are starting to take hold.  The competition among the regional web hosts leads them to expand their offerings and, at the same time, websites are struggling to differentiate themselves.  While he majority of sites are still static content, many require a small backend database.  MySQL is the ONLY FREE database around.  MySQL becomes the default install in many of these webhosting companies.  As tools for building sites become more popular (content management systems, forums, etc.) MySQL is the only viable choice.  (Actually, Postgres is a viable alternative, but not as well known at this point because there is no company pushing it.)

And, in reality, the rest is history.  MySQL has solidified itself as the de facto standard for the web/edge tier in read-only applications.  The LAMP stack is set.

Jump forward to today, however, and the waters become a little muddy.  Many websites do more than hold static data or perform simple operations.  The web is now a tool for doing real business. 

MySQL, who for many years scoffed at the idea of transactions, is struggling to add features to compete.  The big three players (Oracle, SQL Server, DB2) have added low end free versions and PostgreSQL and EnterpriseDB are gaining in popularity every day. 

MySQL is now finding itself in a not-so-enviable position.  They have spent so many years defining themselves in the web/edge space as the best read-only database, that they may have done too good of a job and they may be stuck there forever.

And many questions remain. 

  • Will MySQL ever be able to develop the features necessary to play with the big boys? 
  • Will they be able to overcome the conventional wisdom that they are not an enterprise-class database?

Time will tell... 

Just look at some of their showcase customers.  Sabre (Travelocity) is a great case study for MySQL But, even there, they are stuck in the read-only space.  Sabre uses MySQL on their Travelocity site whenever you search for a hotel or flight.  But, as soon as you decide to purchase, you are immediately switched over to Oracle to handle the transaction.

And I don't want to hear about Google.  They don't count.  Google employs more DBAs, developers and architects than just about anyone else in the world.  They can make any database do anything they want it to.

MySQL - The World's Most Popular Database for read-only, lightweight websites.

Why you need lead scoring

In the world of B2B marketing, just about everything happens "on the Net" these days.  More and more trade magazines are going all digital and the number of e-alerts and e-newsletters I receive on a daily basis is now over 100.  Getting someone to click on your ad or email and come to your site is most definitely more art than science.

But, once a prospect gets to your site, how do you know whether they are really interested without having inside sales call every single one of them?  You need to have a serious focus on lead scoring.

What is lead scoring you ask?  It is ability to rank the quality of a lead based on a concise and quantitative set of variables. 

Notice that I said "quantitative".  The world of open source marketing is, in many ways, not unlike consumer marketing.  Because of the low entry fees, the goal of open source marketers is to get as many leads as possible just like a consumer website.  But, the comparison ends there, because converting a lead to a sale in consumer marketing is relatively easily.  For open source companies there is a fair bit of education that must often be done to counteract the current conventional wisdom (read FUD) or lack of education.  The sales process then looks more like a traditional enterprise sale.

For the tried and true enterprise sales folks, the goal is less about quantity and more on quality because an enterprise sale can take 6-9 months or more.  So, quality is very important.  Open source straddles the fence.  Quantity is key in lead generation and quality is important to convert the lead.

That is where lead scoring comes into play.  Assume you get 10000 leads per week.  You don't have the ability to triage all 10000 leads on a qualitative basis, so you must develop a process to measure each lead based on the information at hand and do it automatically.

How do you measure this?  First, you need to make sure you have the right tools to do it.  Good marketing software, like Eloqua or Manticore, can help.  You also need a great CRM package to hold the information and deliver it in an organized and timely fashion to the right salesperson.

Next, you need to come up with the "what", meaning what are you going to measure?

I recommend establishing a point system.  Leads get points based on both fixed and variable data.

Fixed data includes their company name, their email address and their title.  For example, leads from Fortune 500 companies might get 5 points while independent consultants get 1 point.  Likewise, a corporate email address gets 5 points while an email address from Hotmail gets 1 point.  Why?  Because someone who is just browsing typically doesn't want to hear from you and therefor uses a "spam" address while someone who has an active project wants to keep you top of mind and make sure they know what's up, so they use their work address.

Variable data is more important than the fixed data and often more difficult to track without the right tools. 

Variable data includes: 

  • Number of page views
  • Length of visit
  • Product download
  • Participation in online forums*

*Do they participate in your forums?  If so, that moves them right to the head of the line because it implies that they are using your product and might be getting ready to purchase.

Determine the significance of each "action" and assign points for each of them and mix it with the value of the fixed data.  Then assign a threshold, so when a lead reaches a certain number of points, they are flagged and passed to the right person within the organization to follow-up.

It takes a lot of tweaking over time and you may never get it right.  But, this allows you to apply yet another bit of science to what is, in many ways, art.

Of course, there are exceptions for all of these.  After all, that independent consultant I mentioned earlier may be on a major project with Proctor & Gamble.  And I am not saying ignore the lesser leads all together.  I am just advocating that you rank them so the "hot" leads get the attention they deserve WHEN they deserve to get it.  And, if you are really good at collecting the variable data and passing it to sales, the salesperson already knows WHAT the prospect's problem is and what he/she needs.

The World's Most Popular Open Source Database?

That is the tagline for MySQL and everyone just assumes it to be true.  Meanwhile, PostgreSQL's tagline is The World's Most Advanced Open Source Database and everyone just assumes that to be true as well.  

So, if you say it, is it true?

If I were to say I was The World's Most Prolific Blogger, people would laugh.  So, you can't just say it.  I think you need to prove it somehow, even if anecdotally. 

So, there is a lot of proof out there for both statements.  I have always maintained that:

the best Postgres customer is one who already uses MySQL

Why?  Because they are already suffering with its lack of data integrity, lack of scalability, etc.  Don't take my word for it.  Do a google search or see einhverfr's reivews of both Postgres and MySQL on Ohloh.

Postgres is by far the world's most advanced open source database.  The proof is everywhere.

And I hate to be the one to buck conventional wisdom (right.... lol), but is MySQL really the world's most popular open source database?

I don't know anymore.  There is plenty of proof out there for sure.  But, take a look at freshmeat and you really have to wonder.  Postgres is the 21st most popular project.  MySQL is 336th!!!

Add to that the fact that there is relatively zero marketing and no company behind Postgres and you start to wonder what the real results are.  MySQL tracks customers and downloads.  Postgres does not.  We only know who the Postgres users are if they tell us. 

I think the claim by MySQL may be true only because we can't prove it wrong.  Wouldn't it be ironic if Postgres was BOTH the most advanced AND the most popular?

My next blog post will be about what makes MySQL so popular.

Book Review

At long last, the EnterpriseDB book is available!!!   Woo hoo!  The book was written by Oracle ACE, Lewis Cunningham.  You should really read his blog on IT Toolbox, "An Expert's Guide to Oracle Technology".

Anyway, the book is fantastic. 

I have read through much of the book now and find it to be very educational. Besides the first chapter, which provides some history and color around EnterpriseDB and PostgreSQL, every page contains either code snippets or screen captures.

It is very detailed and specific and any DBA or developer should be able to migrate an application from Oracle or develop a new application from scratch with this book's assistance.
Though you could read it from cover to cover, it is even better as a reference book to keep handy when migrating an app or developing a new app.

Every Oracle developer and DBA should have a copy of this book on their desk and get familiar with EnterpriseDB.

The only downside is that it is based on version 8.1 of the product. Version 8.3 is now available, so hopefully there is an updated version on the way!

Amazon shows the book as not yet available.  But, it will be soon!  I have over 1000 copies here.  If you want one, let me know. 

2007 Year in Review

2007 was a busy year for the database market, which to some people may sound surprising.  After all, the database market is supposedly the most mature and stagnant market.  Even Gartner stopped doing a Magic Quadrant for the database market years ago because there is nothing new...  hmmm..

So, what interesting things happened this year?  Here are a few, in no particular order, along with my commentary

  1. Oracle quietly changed their per-core licensing for Standard Edition and Standard Edition One to per socket pricing.  They claim it was in response to SQL Server, but we all know it's because of the increasing pressure it feels from open source databases like Postgres and MySQL.
  2. MySQL hinted that they would go public before the end of 2007... and then they didn't.  Methinks that they might have been trying to put the heat on Oracle to buy them and wanted to play hardball.  No one stepped up to the plate.  I think they might actually do it in 2008.
  3. Oracle launched Database 11g in July.  Everyone yawned.  And then realized that all the new cool features are add-ons on top of the "arm AND a leg" pricing they charge today.
  4. EnterpriseDB immediately beat Oracle 11g for Best Database Award at LinuxWorld Expo. - A little self-serving?  Absolutely!  It's my blog!!!
  5. Vertica came out of stealth mode...  Didn't you notice?  Maybe you have to look sideways to see them.
  6. FTD Runs Oracle Reports on EnterpriseDB - This story and the  subsequent webcast really woke up the Oracle folks.  google Suddenly, there are 3 Oracle product managers on every EnterpriseDB webcast AND Oracle started buying our trade name in Google Adwords.  Is someone scared?  They should be!
  7. IBM buys SolidDB to compete with Oracle TimesTen In-Memory DB.  This is interesting to me, not so much because of the competition with Oracle, but man oh man, what is MySQL going to do?  I would hate to be an investor right now. 

So, that's seven items I thought were interesting in 2007.  2008 is already heating up to be a fantastic year.  I can't wait to see how it shakes out with all the new things coming...

Catching Up on OLD News - SolidDB

Well, another year has passed and I plan to write a quick year-in-review post, but I wanted to highlight a particularly interesting piece of news from the week prior to Christmas.

IBM announced the acquisition of SolidDB.  IBM has done this, according to their public statements to better compete with the Oracle Times Ten in-memory database, but it once again shows the fragility of the MySQL product set.

In the last 2 years, MySQL has seen its most robust storage engines bought by would-be competitors.  In late 2005, Oracle bought InnoDB who provided MySQL with the only real transactional storage engine. 

Now, IBM has purchased SolidDB, who provided the only other real transactional storage engine for MySQL.

This means that MySQL may have to rely heavily on their yet-to-be-released Falcon storage engine.  Word is that it will become GA with MySQL 6.0. 

If you had a choice between a brand spankin new storage engine or one that has already been proven, what would you choose?

Happy New Year!

2007 is officially behind us and I have been off on a not-so-restful vacation since the 22nd of December.  Tomorrow is back to work for me and I am really looking forward to getting back in the office and getting to work.  Being at home and unplugged from work was nice, but I am ready to start attacking the hill again!

Looking forward to talking to you in the new year and if you are looking for a database, you know where to go ;) 

ENTERPRISEDB!!!