Planet MicroISV - Blogs for and about MicroISVs

A Shareware Life: Interesting Article about Solitaire

Fri, 16 May 2008 18:14:29 +0000

Why We Can't Stop Playing Computer Solitaire

And a related page: On the Profundity of Solitaire

Alex (JitBit Software): Microsoft: the future of medicine

Fri, 16 May 2008 08:05:00 +0000

Video from Mix08. Looks really impressive.

Phil Wright (Component Factory): Sneak Peak

Fri, 16 May 2008 08:01:36 +0000

The upcoming 2.8.5 release of Krypton will contain lots of bug fixes but also a bonus. I am in the process of adding a new built-in palette. It is inspired by Vista as well as Windows Media Player. Here is a peak…

Dharmesh Shah (OnStartups): Insanity? Why A Bootstrap Entrepreneur Raised $17 Million in Venture Funding

Fri, 16 May 2008 01:44:00 +0000

If you've been following OnStartups.com for any period of time, you likely know that I'm not a big advocate of startup founders going out and trying to raise venture funding in the early stages.

My argument boils down to two things:

1) Most folks don't need venture funding in the early stages

2) the odds of first-time entrepreneurs actually raising VC is pretty low.

Oh, and 3) it's one of the least fun activities an entrepreneur can take. Raising funding is often harder than building a product/business -- and much less fun!

So, given my general disposition, it will come as a surprise to many that know me that my startup, HubSpot, announced today that it has closed a Series B round of funding of $12 million. This is in addition to the $5 million Series A funding we raised less than a year ago. So, the total capital raised is now over $17 million. The news was big enough that TechCrunch wrote about HubSpot today.

So, back to the question. Why would a seemingly reasonable and modestly successful bootstrap entrepreneur raise venture funding of this magnitude?

Insanity? Maybe...

How A Bootstrap Entrepreneur Winds Up Raising $17 Million In Venture Funding

1. I seed funded HubSpot with $500,000. To do this, I used some of the proceeds from the sale of my prior startup (which I had bootstrapped with $10,000). The seed funding was an easy decision, because I mostly had to convince myself, and I'm pretty convincing when I talk to myself.

2. The seed funding was enough to build our SaaS product for internet marketing and get it out into the market (i.e. start charging real companies real money to use to it). People bought it. Sure, the product was "pre-alpha" and crappy, but it was useful. We also improved it every day (literally) so it got less and less crappy over time. More and more people bought. This gave us some evidence that there was actually some sort of market demand out there. Interesting.

3. The fact that things were headed in the right direction led us to raise another $1 million in angel funding. For us, that was a fair amount of money (we're capital efficient). Raising the angel funding was reasonably efficient because we had the inside track. The fact we had paying customers was helpful. So, no we're up to $1.5 million in capital raised. Cash in the bank. Life is good.

4. Then, the VC community starts to get interested in HubSpot (this is weeks after we have our angel funding finalized). "Not really interested," we say. We've got a $1 million of fresh cash in the bank. We don't need VC money. As it turns out, one of the best times to raise venture funding is when you don't need the money.

5. My co-founder, Brian Halligan and I have lots of interesting discussion and debate. We'd both debated the whole VC thing while grad students at MIT (where we met). For HubSpot, we had confidence that the market opportunity was big enough to warrant venture-funding, we just didn't think we needed it quite yet. (This is June-ish of 2007). But, we knew we were on to a potentially really big idea. We'd both made some money and weren't really looking for a "modest outcome". We wanted a big, significaint, immodest outcome. So, on the VC front, we figured with the right set of terms and the right partner, we'd consider raising it sooner rather than later. We got the right set of terms and the right partner. So, we raised another $4 million in VC bringing our "Series A" to $5 million. We're off to the races.

6. We did what I think is the best possible thing a startup can do with lots of cash: Not spend it too quickly. No advertising, no marketing, no high-flying salaries for high-flying executives. We hired the smartest, most passionate people we could find. People we knew and respected immensely. People fanatically focused on building a real business and who were constitutionally incapable of spending money willy-nilly. We behaved a lot like we were spending our own money. Because, we were. [Note to self: Write a future article about why VC money is as much yours, once you've given up the equity to get it].

7. Life is good. Sales are ramping steadily. Every month is a record month. Not in terms of visitors, eyeballs or some other proxy for future revenues, but in terms of actual revenues. The business is growing fast. By the time we officially launch the product in November, 2007, we have 100+ paying customers.

8. As it turns out, success attracts more capital. We started getting some "pre-emptive" interest in the company from VCs. "We don't need more money right now," we say. We hadn't even spent half of the last round and lots of cash in the bank. But, we're practical guys and willing to listen. As it turns out, one of the best times to raise venture funding is when you don't need the cash... (see point #4 above).

Fast-forward to today: We've closed a $12 million Series B round.

But, seriously, why did I raise VC funding? Did I change my mind?

The simple answer is no, I have not changed my mind on VC. I still don't think most early-stage entrepreneurs should go out on the venture fund-raising circuit. They should maintain the option of a modest exit. Focus on solving the customer's problem (not the VC's problem). My situation with HubSpot was special. I had already done the bootstrap thing (multiple times) and made money. I had above average odds of raising money for HubSpot.

So, why did I raise funding? Because, this time around I wanted to take a shot at the big leagues. Sure, any success (even a modest one) is nice. But a modest success is not going to change my life much at this point. I want to swing hard. It's not about the money. It's about the fun and excitement of pursuing a really big idea, working with really smart people and doing what I love. [And, of course, the money won't hurt either]

And that, my friends, is why I raised $17 million in venture funding.

If you have questions, feel free to ask them. I'll do my best (within reason) to answer them. Otherwise, I'll keep you posted with future articles as things progress.

Did you like this article? If so, please share it with others.

digg it | reddit | del.icio.us | StumbleUpon

Thanks for your support.
Original Article
Copyright 2007, Dharmesh Shah - OnStartups - Software Startup Blog

Paul Lefebvre (Software Made Simple): Simple is More

Thu, 15 May 2008 14:37:28 +0000

Things that are simple are:

Used more
More useful
More fun
Bring you more customers
Get you more sales

So you can go on more vacations!

Alex (JitBit Software): Tucows forever

Thu, 15 May 2008 12:02:00 +0000

Just received a message from Tucows, a formerly popular software archive. The message says they're accepting our RSS Editor software. Which I've submitted... about two years (!) ago.

Wow. Now I know how a woman feels when her drunk ex-boyfriend calls at 3:00am saying he misses her... While she's trying to figure out who's calling.

PS. It's hard to find the time for full blog posts these days, so follow us on Twitter if you want.

Andrew Duffy: Powerset

Thu, 15 May 2008 09:16:08 +0000

A new search engine called Powerset has been getting a lot of coverage lately. The approach it takes to searching is a nice one: natural language queries seem to be supported in a more advanced way than by stripping out stop words (compare "what is the melting point of iron" to "melting point iron"), and the presentation of results for short queries is very nice (to stick with the theme, see "iron"). The "Factz" are fun – the software attempts to create an outline of a document found in a search by extracting subject-verb-object triples from the text. I’m not sure this works particularly well.

However, the demo only searches Wikipedia and Freebase. It’s very easy to return relevant results when the majority of documents in your collection are about exactly one topic each and are not designed by their authors to cheat search engines. There have been a few attempts to take advantage of this before; for example Turbo 10 federates searches over a number of academic document collections that are not accessible to web spiders like Google and is worth a look if you are researching a very specific topic. I’m interested in seeing how Powerset handles the full web, but in the mean time it's a better Wikipedia search than Wikipedia's own.

Stephane Grenier (LandlordMax): How to Generate Traffic to Your Website - Mark Gladding Review

Thu, 15 May 2008 00:09:10 +0000

Mark Gladding founder of the company Tumbywood Software which sells the software Text2Go, a nice little program that will convert almost any text to audio, reviewed my ebook How to Generate Traffic to Your Website. You can read Mark’s full review of the ebook by clicking here.

Some of the highlights of his review include:

The book is based on Stephane’s experiences over the last few years of marketing and promoting his software company online. This makes the content highly relevant to myself, as I strive to improve the marketing of my own software product Text2Go. I found the examples and statistics quoted in the book fascinating.

The material would also be relevant to anyone starting a small online business, not just a software business. I’m sure the issues are the same. Limited time, little to no marketing budget and 100 tasks all competing for your attention.

Stephane covers an impression range of traffic generation techniques in his book, the highlights being SEO, content generation, freebies, blogging, Google Adwords, press releases and social networking.

I found each topic was covered to just the right level of detail. As you can imagine, separate books could easily be written about each of the above topics. However when running a small business you don’t have the time to become an expert on every possible online marketing strategy. Stephane provides enough information on each topic to get results. Each of the major topics also includes excellent references to more indepth sources of information.

Stephane’s writing style is clear and easy to follow. The writing is illustrated with plenty of interesting graphs and screenshots.

He concludes the review with:

In conclusion, How to Generate Traffic to Your Website contains a wealth of really useful information that’s well organised and clearly presented. I can wholeheartedly recommend it to other small online software and non-software business owners. At $28.95 (or a couple of hundred Google PPCs) the ROI will be very quick.

Something else that’s especially interesting (this is not in his review but was communicated through emails) is that Mark converted the ebook to speech using his software Text2Go and listened to it on his iPod commuting to and from work. How great is that!!!

Casey Software: Software Development Education

Wed, 14 May 2008 18:56:33 +0000

If you're not in the US college/university system, for clarification purposes, a "senior" is someone who is in their fourth and final year while a "junior" is someone in their third year. Got it? Let's go...

On 02 May, I had the opportunity to return to my alma mater - Rose-Hulman - and speak to a class on Software QA. It was a great opportunity and discussion spurred by (hopefully funny) war stories and major screwups that I was more than happy to share.

The most interesting aspect that I saw was the injection of strong Project Management techniques into the curriculum. I don't just mean "here's a deadline... go!". I mean real specification writing, task estimates, milestones, code reviews, deliverables, QA, and overall management. In fact, they get a bit creative in how they do it and even study things such as team psychology and touch on Meyers Brigg. Whoa.

But then they tie it back to the core curriculum...

Part of the fear of every Rose student is their Senior Design Project. Most Senior Design Projects span the entire year, are for real-world companies, are worth 10-12 credit hours (out of 48) and are basically a make or break deal. The way the graduation credits work out... if you don't pass your project, you could feasibly come back for another year.

So how do these fit together?

In order to prepare the students for their upcoming Design Project, Rose starts pounding on them early. All of Junior year is dedicated to putting the pieces in place... such as teaching Code Reviews. So where do they get the code? From the seniors!

Each Senior Design Project gets a set of Juniors attached to them. The Juniors provide support and an additional set of eyes into the project while they get a view into the Senior Design Projects before they have to dive into them. It's a win-win for both sides.

The more interesting part is that it puts some friendly cooperation and peer pressure in place. They've managed to turn both groups into project stakeholders who want and need to find the issues early and while they're as fixable as possible. Failure to do so makes both groups suffer... One anonymous commentor summed it up this way:

The main metaphor that I will remember was that of titanic bugs. This concept is to notice issues weeks beforehand in order to fix them, rather than just days before release when there isn’t much that can be done.

As horribly biased as I am towards Rose-Hulman, it's still good to see somewhere that gets it.

I also wrote about this presentation on the WhyGoSolo Blog.

Alex (JitBit Software): Piracy Concerns

Wed, 14 May 2008 15:00:00 +0000

Piracy is what every software company worries about. We all do our best to protect our software from cracking, patching and other kinds or reverse engineering. But should we really try to build an invincible protection? With all these hardware-lockers, network activations and stuff?

No way. That's what we think here at Jitbit Software. And here is why:

There are three kinds of users:

Ones that will buy your software and never use a pirated version.
Ones that will never buy your software and search for a crack till death. If there's no crack, they turn to your competitor or even buy it with a stolen credit card, which is even worse, because you will have to deal with chargebacks and bank penalties.
Ones that will try to hack (or search for a pirated version), and if it cannot be done easily, they buy (bingo).

Your software protection system should turn group 3 into buyers. Period. A simple asynchronously (to prevent keygens) crypted serial number will do. But if you make it unbreakable, you will have to deal with carders and chargebacks from group 2. If you make it too complicated (network activation, hardware-binding, USB-keys and similar crap), you will lose your customers from group 1.

But this is not the whole story.

We all know, that you can buy a fake Rolex for 40 dollars, or a D&G shirt for 20... That's something we should learn from non-software (tangible) companies: D&G does not fight piracy! Actually D&G even encourages piracy as it promotes the original. That's why when you release the first version of the software, you should use an intentionally weak protection system.

When we released the first versions of our Network Settings Switcher and Macro Recorder back in 2004 (and Network Sniffer later), our serial number system was so lame, that a keygen was out two days after the release. We had gigabytes of traffic and thousands of visitors coming from piracy websites, but we've gained publicity, backlinks, downloads and Google-PR.

Neil Davidson: Remove bottlenecks from your code: ANTS Profiler 4 early access program

Wed, 14 May 2008 10:35:28 +0000

A few months ago we bit the bullet and decided to re-write ANTS Profiler, Red Gate's .net code profiling application. ANTS Profiler 3 was good, but AP4 is going to be remarkable. It's faster (between 5 and 30 times, depending on what you're doing); easier to use (we interviewed and videoed tens of users, and gathered data from hundreds, if not thousands, while designing AP4); and has a great timeline feature (you can highlight areas of high cpu and just see the code that was executed in those time slots).

We've now got a public build available. It's still a very early version but there's enough there for it to be useful. We're going to be releasing weekly builds from now on, so keep an eye out.

If you're writing .net code and want to find bottlenecks in your application, then either read Bart's post for some more information, or go straight to the messageboards to download the early build of ANTS Profiler 4.

Casey Software: 2008 DCPHP Conference is Coming!

Tue, 13 May 2008 16:11:25 +0000

Can you feel it? It's coming...

Even after the glory that was SocialDevCamp, we have another DC local conference coming:

The 2008 DCPHP Conference

This is a real conference which means the agenda is set, the speakers have been determined, and the sponsors are coming.

The highlights:

Christopher Jones from Oracle as one of the keynotes - Oracle has been making inroads in the Open Source community with their Express Edition specifically targetted at Open Source projects and small companies;
Kshemendra Paul of the Office of Management and Budget... if you're a political-insider or a tech-type within the Fed, you know who OMB is and should be excited to see them here;
I'm moderator a panel discussion on PHP IDE's... on the panel, we have Cal Evans from Zend (Zend IDE), Eli White from Digg (Textmate), David Sklar from Ning (Emacs), Jeff Griffiths from ActiveState (Komodo), and Wez Furlong from Message Systems (Vi/Vim).
- If you have questions that you'd like asked at the panel, please contact me - keith at caseysoftware dot com - and I'll see what we can do. There will also be an opportunity to submit questions at the conference.
And of course, we have a plethora of PHP Allstars coming in to speak: Chris Shiflett, Tony Bibbs, Ben Ramsey, Michael Ho, and Damien Seguy in addition to a flock of locals: Barry Austin, Zvi Band, John Croston, and Luke Welling. In addition, somewhere a mistake was made and I snuck in as a speaker. Don't hold it against them. ;)

The sponsors include: A little company call Sun Microsystems... who just bought MySQL, Oracle, and a little company but still a huge champion in and of the DC Tech Community: JobMatchbox*.

So there are lots ofnew and old faces... lots of things to be discussed... and probably more information shared than I'll ever be able to absorb. That's the best and worst part about these conferences. Too much great material, not enough time. But I haven't regretted attending one yet... ;)

* Personally, I'm biased towards JobMatchbox because the owners set up the Panel at American University last month, are the first people I refer job-seekers to, and the dynamic duo who run the place have made a friend welcome in the community. All good in my book.

Neil Davidson: Seth Godin action figure photo competition - an update

Tue, 13 May 2008 11:04:35 +0000

Last week I kicked off a Seth Godin action figure photo competition. I'll keep it open for a few more days and announce the winner early next week. It's not too late to enter.

The entries are so good I think I may well have to hand out several prizes. The weirdest photo category will be particularly hard fought.

So far Jim Kring, Tom Randle, Mr Flibble, embe and s3a have all submitted photos.

If you want to see Seth being attacked by a teddy bear, upside down in a glass of cider, posing as Lenin, reading his little book of marketing secrets or tied up as Darth Vader's prisoner then check out the photos on the original competion post.

Ian Landsman: Follow me on Twitter

Mon, 12 May 2008 18:45:34 +0000

Just an FYI, if you haven’t subscribed to me on Twitter please do. I still write here from time to time, but it’s hard to find the time for full blog posts these days. The low commitment level of Twitter allows me to normally post several times a day. You’ll also get more details about ongoing HelpSpot development than here as I sometimes Twit about what I’m working on right then.

Casey Software: SocialDevCamp East - Recap

Mon, 12 May 2008 15:44:52 +0000

I feel like it's all been said already, but since that's never stopped me from talking before... ;)

Wow, SocialDevCamp was amazing.

First, the opening were great. Ken Yeung was good enough to catch some of the opening discussion on Viddler. Look past the chaos for a minute and see how order appears and negotiations happen implicity. It's amazing how when you have a lot of smart people around and competing interests that we can work something out without anyone being *too* disappointed. We started with 25+ sessions and were able to filter it down to 16 pretty easily. We had 4 sessions going concurrently throughout.

For sessions, I attended:

a detailed discussion on Amazon EC2 and S3. Jimmy Garnder - the creator of MyDropBin, a simple S3 front end - was in attendance in addition to numerous EC2 users doing some advanced things. The ideas, information, and usage kicked around was pretty impressive.
I popped in and out of the second session, but luckily Ken also caught that on video: Social Media & Semantic Web from Dr. harry Chen. The bits and pieces that I caught were fascinating and touched on Taxonomies, Ontologies, and other things with names. ;)
I hosted the third session of the day on "Building the Social Hub" where we talked about integrating with third-party platforms. This was an interesting one to follow the Semantic Web discussion because Bear of Seesmic shared all kinds of tips, tricks, and ideas to integrate things indirectly via semantic web concepts.

The final discussion was hosted by Jared Goralnick of AwayFind (sponsor) and discussed the idea of the DC-Baltimore-Philly tech corridor and deserves a discussion of its own. We talked details about what makes us different than Silicon Valley, how we can benefit and utilize those differences, how we can close the gap on others and generally spark things moving faster. One of the single biggest differences that strikes me - and I noted it during the discussion - is the DC Tech Scene's view of competition. Yes, we compete, but we're all connected too. We don't tend to compete against each other... we're working together and fighting the same fights, so we cooperate and coordinate. I don't see that in Silicon Valley or hear it from my friends there. It's a different idea. After the main discussion, Clarence Wooten of CollectiveX shared his term for it: cooper-tition. Mario Armstong of NPR/XM shared some experiences working with O'Malley when he was still Mayor of Baltimore.

Afterwards the festivities at Brewers' Art were fantastic. We ended up with just a little more funds than we expected, so towards the end, the food ended up being on the same tab. I also shared some tips with Mario on some ideas for using Social Networks for competitive research. I've shared some of those ideas in this space... some are staying off the record. ;)

And finally, Dave & Jennifer Troy, Ann Bernard, and myself acted as the organizers/cat herders of the operation. While I've received dozens of emails, calls, im's, and dm's thanking me/us for everything, but I think a good portion of that is misplaced. Yes, Dave found the venue. Yes, we all contacted sponsors. Yes, we all got the word out and pushed people to show up. Yes, Ann coordinated presentation selection and handled... announcements. ;) But it's leaving out the biggest part...

The participants.... Without the people who showed up, who volunteered, who volunteered to speak, who asked questions, who gave their opinions, who shared drinks after, who kicked around and tore apart ideas... the whole thing wouldn't have worked. That's what an unconference is. It's self-organizing and the good bits and information float to the top and the real experts and opinions filter out. That's what happens.

And if you don't believe me on how much people enjoyed it... check out the generous coverage:

Jimmy Gardner (sponsor), Shashi B of Network Solutions (sponsor), Ann Bernard (organizer) on WhyGoSolo and her own blog, Dave Troy (organizer), Ken Yeung, and Greg Cangialosi (sponsor). There's a larger sample of the coverage in Twitter via Hashtags and Summarize.

For a more visual review, check out SocialDevCamp on Flickr, Jimmy's photo collage, Mark Scrimshire's photo collage or maybe the videos from Ken Yeung or Dave Troy.

The final - and my favorite video - is from Shashi B. Yes, I'm in it and the first half is dedicated to WhyGoSolo, but the second half is key. In the second half, I share how we managed to collect the last 20% of the funds to pull off SocialDevCamp East. If you ever doubt that one of these events are community driven, listen to that half of the video and think about it. This unconference happened because of the community, because of the participants, and we - Dave, Ann, and I - were simply lining people up in the same direction.

Neil Davidson: Interview between Peter Day and Michael Moritz

Mon, 12 May 2008 14:33:50 +0000

I heard this interview with Michael Moritz, the Sequoia Capital venture capitalist, on Radio 4 the other day. It's well worth 30 minutes of your time. Moritz talks about his journey from childhood in Wales, history degree at Oxford, journalist for Time magazine and on to Sequoia where he has invested in, among other companies, Google, Apple, Yahoo!, PayPal and Cisco.

Here's an excerpt (apologies for the ropey and inaccurate transcript):

Peter Day: Many investors throw their money at lots of investment targets, hope that two or three will be survivors and that one will be a superstar success.

Michael Moritz: That isn't how things are done at Sequoia. It's not the way you think of it. Every single time you write a cheque you expect, or pray, depending on your inclination, for that investment to succeed.

You can hear the interview as .ram streaming audio, or a downloadable mp3 (note the interview doesn't start until about a minute in). The Radio 4 web page is here.

Eric Sink: Upcoming Gigs

Mon, 12 May 2008 14:19:41 +0000

In July I will be giving a keynote address at GUADEC, the annual GNOME conference, being held this year in Istanbul.

In September I will be speaking again at the Business of Software conference, being held this year in Boston.

And finally, for something completely different, don't miss the Jam Session at Tech-Ed on June 3^rd. Several of us minions from SourceGear are planning to take the stage and give our rendition of Pinball Wizard. It'll be me on acoustic guitar, our development manager Jeremy Sheeley on bass, and our product manager Paul Roub playing the Evil Mastermind Schecter PT that will be given away later that week.

And BTW, none of us will be dressed as The Evil Mastermind. This should be obvious, as The Evil Mastermind would never do something actually cool like a song by The Who. Rather, he would do something like a Kelly Clarkson song and mistakenly believe it was cool. :-)

Tim Haughton: Amazon SimpleDB

Mon, 12 May 2008 14:00:06 +0000

Amazon's SimpleDB service is a reliable storage service for small pieces of textual data. It provides the opportunity for the storage, modification and retrieval of datasets, but without the need for the maintaining of a more traditional database server, which should be a major benefit for a thinly spread micro ISV. It should be noted, however that the SimpleDB service is just that; simple. It does not have anything like the bells and whistles of a full RDBMS database.

Overview

Like the other Amazon Web Services, SimpleDB has a limited number of entities:

Domains. A domain is what might be considered the equivalent of a database instance. Just as you might create several databases on a single RDBMS server, you can use domains to partition logically distinct datasets.
Items. An item is a uniquely (within the domain) named collection of attributes that represents a data object. You can add, modify or delete an entire object in one go, or modify individual attributes.
Attributes. An attribute is a uniquely named (within the item) category of information.

The main difference between SimpleDB and a full RDBMS is that information is stored in hierarchical trees, not in tables. There are also no predefined table schemas, so any item can have a different set of attributes to any other item in the domain. Whilst this provides enormous flexibility, you will need to keep your wits about you, as if you misspell an attribute name, you might find that that particular nugget of information is lost forever.

There are no types in SimpleDB, other than the string. All information is stored as text. This means that SimpleDB can only perform case sensitive string comparisons. There are no integers, floating point numbers, dates etc. Again, it's simplicity and flexibility also means that the SimpleDB system won't even raise an eyebrow should you provide, say, a floating point number for an attribute that should contain dates. Maintaining the integrity of your data becomes very much your concern.

The SimpleDB query language is very limited. It is far simple than SQL, for example. Queries take the form of

['AttributeName' Operator 'Literal']

In addition to the Boolean operators NOT, AND and OR, you have the equality operators, =, !=, <, <=, >, >= and starts-with. So to find items with a membership expiry date before the end of the year, we would run this query:

['MembershipEndDate' < '2007-07-01']

Note that this is just a string comparison, you must encode your data in such a way that makes for sensible comparisons.

Like with Amazon's other distributed systems, applications using SimpleDB will need to take into account the propagation latency. At any one time, one cannot be sure that the data in the domain are not out of date. In reality, it takes only a few seconds for all of the physical servers to achieve consistency, but any application that requires constant integrity will need to employ some sort of caching mechanism.

Attributes in SimpleDB are severely limited in size, specifically, to 1024 bytes. This limitation is entirely deliberate. It is intended that objects of any appreciable size will be stored in S3 rather than SimpleDB. You should attempt to work with this rather than work around it. Storage costs in SimpleDB are many times the storage costs of S3.

Pricing

Storage space in SimpleDB is expensive; $1.50 per gigabyte. Your space usage is calculated from every item, attribute name and attribute value in your system. In addition, you are also charged for an additional 45bytes per item, attribute name and attribute value. This is for the indexing that SimpleDB automatically performs.

Data received by SimpleDB is charged at $0.10 per gigabyte, and data sent by SimpleDB is charged on a sliding scale from $0.13 to $0.18 per gigabyte.

You are also charged $0.14 per hour of machine usage. The amount of CPU usage depends on the volume of the data returned by the query, the amount uploaded etc.

Do you use SimpleDB? Does it work for you?

Ben McGaughey (BugNinja): Amazing resource for anti-cracking information

Mon, 12 May 2008 12:00:19 +0000

I've seen some write-ups on preventing software piracy and cracking in your applications, but this one is absolutely amazing:

http://www.inner-smile.com/nocrack.phtml

Ben McGaughey (BugNinja): Required: Hand-Holding on Order Forms

Mon, 12 May 2008 11:41:05 +0000

Steve Friedl wrote this page about credit card forms on eCommerce sites asking users to format their card numbers a certain way. As a tech savvy person, I completely understand where he's coming from. As someone who has done a lot of customer support for eCommerce sites, I think I can provide some feedback as to why this happens so much.

What I really disagree with from the page:
"If there were some security or integrity reason for disallowing these characters, I guess I'd buy it, but I've not found a single good reason for it. The consensus among those that I've spoken to is that it's nothing but lazy, sloppy programming. I completely agree."

Ouch.

I guess he is frustrated that programmers can't put in some simple regex or other string parsing routine to change the format of the numbers, but that's not really the point, it's not about our ability as programmers to do this - it's quite easy, agreed, and though I'm not sure about this, some of those forms he highlighted may actually do the regex after the form is submitted.

It's actually about customer support.

Believe it or not, sales are lost because people don't know if they are supposed to put spaces or dashes when typing their credit card number. You literally have to tell people exactly what to do when it comes to your order forms. Any questions, and people will walk away.

It's absolutely true. It was more true back in 99, but it still happens today. I will get an email from a potential customer on one of the eCommerce websites I help manage, prior to purchasing, asking me whether or not they need to put their middle initial with their name on a credit card form, or if they are supposed to put dashes or spaces in their card number.

Having these "no dashes or spaces" statements and restrictions on credit card forms comes from years of handling customer support issues and then applying that knowledge to the setup of an eCommerce site. After 2 people email with the same question, and after possibly losing 1 or 2 sales, most of us just slap this language on the forms without thinking twice about it.

If someone has to ask themselves even one question about their purchase with you, many times, they simply will not buy.

A personal purchasing story:

I was just about to purchase a great "skinning" component for my apps that made them look very pretty, but during the purchase process, I wasn't sure if I should buy a "single developer" license or a "company" license - all because the language didn't make sense to me. I was only going to use this component on a SINGLE development PC, but I was going to use it within my COMPANY for a commercial product.

I didn't buy the component right away.

I even emailed first to ask which license I should buy.

The email bounced back. Gah. Now they definitely aren't getting my order.

Customer Support Story:

One customer kept entering 10/04, 10-04, 1004, etc for the expiration date on their credit card. They emailed saying it wouldn't let them through to make the order and they've tried every possibility for their expiration date.

Turned out their card expired in September. That would be 09/04 :)

From then on, expiration date fields are made up of 2 drop downs:

Jan-1
Feb-2
Mar-3
etc
/
2008
2009
2010
etc

*************************************

Lastly, think of it this way:

Have you ever stood in line at the grocery store, swiped your debit card, then it said to enter your pin, hit the green OK button, then the Yes button?

Then you go across the street to the hardware store, swipe your debit card, and it wants your pin, hit the third key in from the top to say Yes, then the big OK button, then you are done?

Have you ever noticed that some of these places have little pieces of paper scotch-taped to them to help people understand what to push and when?

It's the same thing online. Everywhere you go it's a little different. We, as vendors, must provide every instruction needed to make it through the buying process because we aren't right there to help them, they need that little message scotch-taped to the order form.

Stephane Grenier (LandlordMax): Web Design for ROI - Review

Mon, 12 May 2008 09:30:31 +0000

I recently got my hands on a copy of the book Web Design for ROI: Turning Browsers into Buyers & Prospects into Leads. Overall I have to say it was a very good book, it really explained a lot of the key concepts in web conversions at a high level (sales conversions, lead conversions, etc.). For example I know our site isn’t perfect, and we already have a lot of work planned towards this end, but what I really liked is that most of the “enhancements” we’re planning for the future are described in this book. At least at a high level. This book is not about the details of the implementation, but it’s a high level description of what needs to be done and why.

Which is the key thing to remember about this book, it’s written at a high level, meaning it’s aimed at people who are newer to the concepts of website conversions (sales, leads, etc.). Therefore if you’re already well seasoned on concepts such as sales conversions it won’t be as exciting for you. It’s not as detailed or thorough as books like Call to Action, but it’s still very good. This book will be especially interesting for you if you’ve been mainly focusing your attention on traffic generation and ignoring any efforts on items such as sales conversions, etc..

I’d highly recommend this book for anyone who has a website where they’d like to increase their conversions, be it sales, leads, etc. It’s a good introduction and will get you going. What I really appreciated was the “Digging Deeper” chapter at the end. The authors really hand picked the best books and websites on the market to recommend as further resources to investigate. Personally I’ve almost read every single book they recommended, as well as follow most of the websites they recommended. Some examples of the books they recommended (which I’ve recommended - even reviewed - before here on my blog) include: Don’t Make Me Think: A Common Sense Approach to Web Usability, Call to Action: Secret Formulas to Improve Online Results, Blink: The Power of Thinking Without Thinking. And some examples of the websites they recommend include: Marketing Sherpa, Guy Kawasaki’s website and blog,

Overall I recommend this book, especially if you’re just starting to get into the market of increasing the performance of your website. It’s not as advanced as some of the other books out there, but that’s not it’s intention. It’s a great book to start with and build up from. Well worth the purchase.

Gavin Bowman: Micro ISV Digest

Mon, 12 May 2008 10:19:33 +0000

A weekly roundup of Micro ISV news, announcements and relevant blog posts.

News and Announcements

Paul launched YesCalculator, a calculator. via BoS.
Give away of the day results on BoS (followup).
BdJ results from Follow Steph.

Relevant Blog Posts and Articles

Why making a fool of yourself is a good thing, from Neil Davidson.
It's the tech guy, obviously, from KC.
Amazon S3 for Micro ISVs, from 47hats.
Also, Information overload tips for Micro ISVs.
7 steps to speed up your technical support, from Dennis Crane.
Notes from MIT Underground 2008, from OnStartups.
The evolution of a software startup, from Instigator.

Further (mostly relevant) Reading

That big project and your relationship, from WWD.
The art of war for entrepreneurs, from Pam Slim.
What do you know? from Seth Godin.
Big Bucks from tiny computing startups, from Information Week.
Aesop's top 7 tips for a happier life, from The Positivity Blog.

Alex (JitBit Software): If you have nothing to blog about...

Mon, 12 May 2008 08:33:00 +0000

...just don't blog.

Phil Wright (Component Factory): Quad Core Overclocking

Mon, 12 May 2008 05:18:34 +0000

Research your Processor

Having never overclocked a PC in my life the first step was to do some research. Just entering overclocking into Google gives plenty of resources for you to read up on the topic. To narrow it down trying entering the name of your processor as well, chances are you will find plenty of people giving reports on how they overclocked your exact processor.

I found several people that had overclocked the QX9650 to around the 4Ghz level without much difficulty. The more serious overclockers have managed 4.5Ghz using just air cooling. While I am not interested in trying to reach such record breaking levels, if there is some free extra processing power available then I might as well grab it.

Overclocking Tools

Getting started means downloading some tools. Start with the excellent CPU-Z utility which gives detailed feedback on your machines setup. Next you need a tool that can be used to stress test the processor and memory to check if your new settings are stable. I used OCCT which was created by a group of French overclockers and is really simple to use. It thrashes all four cores and memory and when it detects a stability problem gives a nice sound effect to alert you.

The bane of all overclockers is the processor temperature, I used SpeedFan to monitor and graph the temp of all four cores. No point in running fast if that expensive silicone is being turned into cheap charcoal.

Baseline

As all good programmers know, measuring performance requires a baseline set of data to use as a point of comparison. Here is the default setup of the hardware via CPU-Z…

To measure the real world performance of the machine I am going to use my build script. There are plenty of synthetic tools to measure the raw performance of memory, processor, graphics and so forth. But as far as I’m concerned I want to see the benefit in real work and not just some theoretical benchmark. Running the build of Krypton game a time slightly over 36 minutes…

Note that I only ran this test just once. To be more accurate I should have run it half a dozen times and taken an average of the different times. But frankly I am not bothered with that level of accuracy and don’t want to spend 3 hours gathering the data.

Updating the BIOS

Doing the actual overclocking is simple. Enter the BIOS at startup and then modify the cpu multiplier. You can see in the CPU-Z picture above that the default multiple of 9 gives a 3GHz speed based on the base 333Mhz speed of the bus. I can increase this in 0.5 increments via te BIOS and so did just that. Adding 0.5 and then running the OCCT stability test for 30 minutes to ensure it was stable and watching SpeedFan to ensure the processor was not cooking.

Repeating this process I managed to get a stable system with a multiple of 10.5 giving a cpu speed of 3.496Ghz with the temp peaking at around 63c. Going above this caused the system to become unstable with regular crashes.

In order to get stable again I increased the voltage to the cpu. Usually this will solve the instability at the cost of running the cpu hotter. Slowly increasing the voltage caused the temp to increase markedly and too much for my liking. Once I was getting 75c I decide to stop as I don’t want to fry a very expensive processor.

Note that I am using the stock Intel cpu fan that comes with the processor. Switching to an aftermarket cooler/fan would probably have reduced the temperature enough to get a higher stable multiple. I did buy a Zalman fan for my machine but it was far to noisey for me to tolerate listening to all day long, so I switched it out for the stock fan before starting the overclocking.

I now changed tack and tried overclocking the bus speed, giving a boost to the FSB and the processor at the same time. This has the added benefit that you can make small increments in the cpu speed rather than having to jump in large increments.

Starting from the default of 333Mhz I upped the value a little at a time until the system become unstable. Then backing up to the last good setting I managed to achieve 340Mhz. This gives a FSB speed of 1360 and ups the processor to 3.570Ghz. You can see the settings here…

This seems to be about the best I can manage. If the cpu is any faster the temperature goes to high for my liking and making the bus faster results in an unstable system. This is someway short of the 4Ghz that others report but the I am using a stock Intel fan for cooling and bog standard DDR2 memory.

Real world performance

But how does this impact actual real world performance? Running our build script we now get just under 31 minutes…

This represents an improvement of 16% which is very close to the actual processor overclocking of 19%. Given the build process includes a fair amount of reading/writing to disk this is very respectable. I then ran the OCCT stability test for 9 hours overnight and it didn’t crash. So I feel confident in leaving these settings alone and running with them all the time. After all, why turn down a free speedup of 500Mhz!

Intel Observations

It seems strange that Intel are releasing processors that seem to be capable of running so much faster. Why would you sell processors rated at 3Ghz when clearly they can run much faster? Obviously they are going to sell them with a safety margin but the difference between the rating and actual maximum speed seems odd.

I don’t remember previous generations from either AMD or Intel having this level of overclocking ability. Are Intel deliberately holding back because they already have a big performance lead over AMD? If AMD start to catch up would Intel suddenly start selling faster rated processors? I wonder.

Keith Alperin: MercuryMover v1.1.2

Sun, 11 May 2008 22:57:52 +0000

Hot out of the Helium Foot Software Forge is MercuryMover v1.1.2. Get it right here. 1.1.2 fixes a long standing latent issue that caused MercuryMover to crash on start after a reboot. Fortunately, this issue only affected Mac OS X 10.4 (Tiger) users. Fortunately (for me) i have some of those users as customers! Unfortunately for those customers, they had to reinstall MercuryMover after each reboot. Although it's downright embarrassing to release a bug like that into the wild, this cloud is not without a silver lining. There were a number of customers that helped me track this bug down by sending in bug reports, console logs and preference files and i always enjoy meeting and corresponding with my customers. ...