This news article made me cup my hands in joy. No, I am not an investor in MySQL, but just that I love to see underdogs doing well. But will the move succeed? A quick look at the VA Software (the folks who created Source Forge) stock does not seem encouraging.
I am keeping my fingers crossed, but here's why I think MYSQL will not go the VA Software way:
· VA Software's flagship product is SourceForge and databases are any day a hotter space than source management. Even the largest of development efforts target a few thousand developers, while the largest DB installations run into millions of users.
· MYSQL is offering an enterprise wide license at just $40000. Seriously, the likes of Oracle and Microsoft have been ripping off customers with ridiculous prices. I don't mean these products are not good. They are great which is why they command a premium. But why pay the earth to "overengineered" Oracle when you can pay a fraction of that to "lean and mean" MySQL, especially when you do not need the bells and whistles? I see the lower end of the corporate market looking at MySQL more seriously.
· Oracle's attempted takeover last year vindicates the model.
· MySQL has successfully piggybacked on several open source products and bundled itself into platforms like Solaris,MacOS and Linux. At the end of the day if a data centric company like Google can rely on MySQL I see no reason why others can't. It is proven!
So, who's joining the IPO party ?
Products Technology
Linking and Sharing
Trackback
Trackback URI for this entry:
http://www.prakashonsoftware.org/blog/index.php/2007/01/31/mysql-to-go-public/trackback/
Wikipedia defines a mashup as
A mashup is a website or application that combines content from more than one source into an integrated experience. It is sometimes created as a critique or commentary on an existing work or product. The tactic owes much to previous recombinant forms including: content repurposing, most notably by Kenneth Anger in his 1963 film Scorpio Rising; DJ mixing and culture jamming.
Mashups are real cool. No doubt about that. Just check out Link Detail. This little mashup marries Google maps and YouTube and lets you view videos from a location by clicking on an area in a map. The obvious question: Can a mashup make money? Here are some possible business models:
Skim and run: A short term market skimming strategy driven by high volumes of focused users driving advertising revenues. Say you have a huge upcoming sports event and somebody puts up a mashup that aggregates relevant content from multiple sports websites in a useful way that has not been done before. The sheer usefulness would driver users to the mashup and the focused user base would ensure advertising revenue.
Enrich and resell: This option is best suited for companies that have rich proprietary data stores that are not of much use standalone, but that have great usefulness when combined with other data. Here data available publicly is enriched with proprietary data and resold. Basically, you wrap and exisiting API, add value and resell.
Scrape and recombine: Good old screen scraping (yeah you have decent API's that help you) can get glamorous if you can combine it with other scrapped bits in useful ways. I cannot think of something concrete, but the appeal lies in the fact that you are not using somebody's API (over which you have no control) but HTML. It seems possible to recombine data in ways that are not easily traceable to the source! This could make the model a tad bit more difficult to ape.
All said and done, I do not see the individual genius mashup developer sitting in his drawing room and making money for the following reasons:
-Data is at the heart of any mashup and someone else owns it. If the data pipeline is cut off the whole model chokes.
-Most mashups do not seem to add any extra proprietary value over and above what public API's provide.
-Too many restrictions on usage that most data/API providers have.
-It is all too easy to imitate and without any proprietary lever, a sustained competitive advantage is infeasible. Any smart developer can put together a competing mashup over a few weekends.
Mashups however can get interesting monetarily if data rich corporates decide to team up. Technology Web2.0
Linking and Sharing
Trackback
Trackback URI for this entry:
http://www.prakashonsoftware.org/blog/index.php/2007/01/29/monetizing-a-mashup/trackback/
This impressive Sandhill.com write up brings out the key realities in today's complex selling environment. My elaborations on these points from a software services context are below.
Importance of homework : Knowing the industry and the company pain areas are invaluable when it comes to establishing connect with CXO's. The more closely your "pitch" relates to the pain areas, the better the connect. In my experience, a lot of money gets spent into chasing industry "fads". The current favorite in the banking industry seems to be SOA. Different people mean different things when they say "SOA", if you want to light up the eyes of a CTO in a bank SOA is a good bet. Taking this to the next level would mean doing homework on the CXO you are selling to. I once interacted with a CXO for whom "XML" was the mother of all solutions. Whether it cross platform communications or UI design standards the guy would look at XML. Speak the three magic letters and it was not tough to make a connection.
Translating the sales message: It is rare for a single stakeholder to make the decision all by himself. It is important to "translate" your message into the "language" spoken by each stakeholder. Most services sales situations have three categories of stakeholders: a).Economic b). Technical c). Business. It is important to have a cross functional pre sales team at onsite to adequately address the concerns of each category.
Broad basing relationships: Good relationships at the operational level are easily undermined by competition that seeks to build relationships at the strategic level. Most services companies focus on operational level connect. I think this is natural given the nature of work. But it is so very important to have at least one person onsite who does nothing but listen to the CXO's, be their trusted advisor and seek to drive the selling decision top down, rather than leave it to the operational folks.
Let the client drive the financial case: The numbers game! ROI and the like. In my personal experience, trying to calculate the ROI that your solution can have for a client is impossible. The details needed are too contextual, political and sensitive. It is easy to fill in numbers into a nice looking excel sheet, but walking the talk is something else. Let the client do the number crunching! sales
Linking and Sharing
Trackback
Trackback URI for this entry:
http://www.prakashonsoftware.org/blog/index.php/2007/01/26/sales-tips-from-sandhillcom/trackback/
Take a trip: Before you sign a contract, spend some time at the offshore outsourcing center, even if it's on the other side of the world. Many companies send business to centers overseas without ever visiting them.
Consider culture: Offshore workers may speak your language, but they may not fit the company culture. Want programmers to challenge ideas and propose alternate courses of action at times? You may be better off in Moscow than Bangalore.
Look beyond the bottom line: Many outsourcing deals are focused solely on cost but break down in the third year after every last penny has been squeezed out. Instead, consider what you'll need over a three- to five-year period and choose a vendor capable of innovating or making enhancements when the time comes.
Seek maturity: The Software Engineering Institute rates the maturity of business processes of some offshore IT providers. Look for providers with CMMI (capability maturity model integration) Level 5, the highest possible rating.
You can read the caveats of outsourcing here.
Outsourcing
Linking and Sharing
Trackback
Trackback URI for this entry:
http://www.prakashonsoftware.org/blog/index.php/2007/01/20/business-weeks-tips-for-outsourcing/trackback/
This is a new one. Companies hiding behind the law and overcharging customers. WSJ points out this $1.99 rip off by Apple. Apple probably got away because their user base is so terribly loyal and will not switch to Wintel come what may.
Technology
Linking and Sharing
Trackback
Trackback URI for this entry:
http://www.prakashonsoftware.org/blog/index.php/2007/01/20/loyal-customers-screwed-over-royally/trackback/
The McKinsey survey on CIO's Agenda for 2007 throws up some interesting ideas. Here are the two key points that CIO's would focus on according to the survey:
- Migration to service-oriented architectures inspite of all the confusion over what "service oriented" really means.
- Introduction of lean-manufacturing principles to data center operations. Applying lean principles can help reduce waste and improve labor productivity by as much as 40% in some processes.
I am not surprised that Web 2.0 does not figure in the list. :-)
Technology
Linking and Sharing
Trackback
Trackback URI for this entry:
http://www.prakashonsoftware.org/blog/index.php/2007/01/18/cio-agenda-2007-thoughts-from-mckinsey-2/trackback/
A lot of debate is currently going on about "Enterprise 2.0". Some authors define it as Web 2.0 applied to the Enterprise, while others take a wider view. Here's why I think most of the touted benefits will remain more hype than reality:
Legacy problems: If any technology is to hit the Enterprise mainstream it is important that the Citibank's and the Walmart's of the world adopt it. But with a huge volume of business rules locked up in dinosaur legacy systems of mainstream companies, widespread adoption will remain a challenge. The so called "Enterprise 2.0" adds nothing new to solve the legacy issues but is just a repackaging to myriad buzzwords with Web 2.0 thrown in.
The easy coupling myth: One of the things touted about is "ease of integration". Haven't we been through this hype cycle before ? XML standards based Web Services and the like ? Systems integration is one problem where there can be no silver bullet. Disparate standards, incompatible workflows, lack of trust between vendors and partners and misaligned goals will always keep "easy coupling" a mirage.
Minimal maintenance myth: Given the teething legacy issues, Enterprise 2.0 software is at best likely to be some sort of wrapper or glue that can tie up a few ends or create a snazzier "user centric" UI for some back end system. The real maintenance overhead is underneath the covers and not in the wrapper. Adding layers is likely to increase your IT infrastructural overheads with more investment in people and processes.
Business is about value not "coolness": The social engineering paradigms underlying several Web 2.0 concepts maybe valuable for certain B2C (Business to consumer) contexts. A look at startupreview.com throws up several interesting Web 2.0 based startups, but this still leaves me wondering: where is the over riding value (that justifies the hype) from an enterprise perspective ? Granted your marketing and HR departments can use blogs and your R&D team might collaborate using a Wiki but apart from being "cool" the business value does not seem to justify the excitement and hype.
While Web 2.0 and Enterprise 2.0 surely have great potential in certain limited contexts within the enterprise, anyone who expects revolutionary change is climbing up the hype curve.
Linking and Sharing
Trackback
Trackback URI for this entry:
http://www.prakashonsoftware.org/blog/index.php/2007/01/18/enterprise-20-big-hype/trackback/
I am sick and tired reading articles and comments that berate the quality of immigrant technical talent. Here's an article that should make every immigrant to the US proud. I will leave you to savour the full article, but will highlight some quick facts that should interest my Indian readers.
- Indians have founded more engineering and technology companies in the US in the past decade than immigrants from the U.K., China, Taiwan and Japan combined.
- Of all immigrant-founded companies, 26% have Indian founders.
- The second largest group of immigrant non-citizen inventors were Indians. The Chinese did us in here.This is becoming a familiar trend.Grrr..but I still love their food. :-)
Linking and Sharing
Trackback
Trackback URI for this entry:
http://www.prakashonsoftware.org/blog/index.php/2007/01/17/the-value-of-the-immigrant/trackback/
Traditionally services companies have focussed on the Deutsche Bank's and the Walmart's of the world. The Sandhill Group points out key differences between the mid market and the large companies in this concise report. Here are the highlights:
-The vast majority in either group are unwilling to outsource strategic functions. No surprise here.The only time I saw a strategic function getting outsourced was when the outsourcer was in the ICU.
-In mid size companies the percentage of CIO's reporting to CEO's is 10% higher than in large companies. Probably indicates that mid size companies are probably more technology driven and are perhaps using IT as a key source of competitive differenciation.
- 72% of Mid size businesses say Business Process Improvement is the top priority, but strangely a whopping 49% don't use or don't plan to use the same. Life is good, if you work with Accenture.
Linking and Sharing
Trackback
Trackback URI for this entry:
http://www.prakashonsoftware.org/blog/index.php/2007/01/16/selling-to-the-mid-market-some-interesting-insights/trackback/
UI design is something that receives less importance compared to the holy grail of coding. But we all hate websites with crappy layouts and cumbersome navigation structures. I came across this really neat tradeoff matrix on UI design by Mike Padilla.
| Design | Benefit | Cost |
|---|
| Shallow information architecture | Fewer clicks to find info | More clutter |
| Deep information architecture | Clean, reduced clutter | More clicks to find info |
| Small font | More information per screen | More difficult to read for some users |
| Large font | Easier to read | Less information per screen |
| Drop-down box | Selection amongst many choices using limited space | Hidden choices |
| Radio buttons | See all selections at all times | Additional space required, clutter |
| Icons | Quick recognition once learned, aesthetically pleasing | Must be learned |
| Text links | Always understood | Must be read, do not stand out as actionable items as much from other text |
| Abbreviations | Save space | Must learn or recognize |
| Full text | Easily understood | Requires additional space |
| Keyboard shortcuts | High speed of data entry | Must be learned |
| Point and click | Intuitive | Additional time required for interaction due to increase motor skills required |
Mike goes on to suggest the following factors to judge a UI:
Ease of learning and memorability Efficiency of use Error frequency, severity, and recovery Subjective satisfaction
programming Technology
Linking and Sharing
Trackback
Trackback URI for this entry:
http://www.prakashonsoftware.org/blog/index.php/2007/01/14/ui-design-choices/trackback/
Tell a friend
