Free version available
Magnifying Glass

Blog archive for August, 2008

Syndicated Posts

FaceBook

Twitter

  • #financialconfession of the day: "I get really annoyed when my friends spend frivolously and then complain about... http://fb.me/I1qhJW4p 4 days ago
  • "Bye Bye Quicken Online, Hello Mint!" by Jay Monee on Budgets Are Sexy - "PocketSmith: This big pull with these... http://fb.me/EuooR2LH 1 week ago
  • "Too many people spend money they haven't earned, to buy things they don't want, to impress people they don't like." Will Smith #quote 1 week ago
  • More updates...

Pixel perfection, design and branding

Sunday, August 24th, 2008 by Jason

Quick post tonight before I head off to bed. James and I are knitting the Beta 1 interface together, and I thought an excerpt from this email to the team worth posting.

I used Time Machine to restore a file for the very first time! Thank god I plugged it in tonight as an afterthought - I accidentally overwrote the gradient png when saving another slice, and as fate would have it I didn’t have an original PSD of the gradient (was a modified long one). So, popping into Time Machine and restoring the file probably saved me 20 minutes.

Fun fact – the event window design alone took me four hours! I used to knock up a whole site design in that time! Goes to show how much attention to detail our baby is getting :-) Kinda wish I’d tracked time on the rest of the design now too, for interest’s sake.

Being a ‘pixel perfect’ designer has taken a little getting used to, especially when coming from an environment where design time was limited. I’ll briefly explain.

For most small-to-medium enterprise websites on a modest budget, I adhere to what I call “the Threshold of Appreciation” – that being, clients that fell into this bracket wouldn’t value the additional amount of time and detail put into their designs anyway, so my strategy was to adopt broad strokes in the application of good design, but not to sweat the tiny details. 

It’s effectively Pareto Theory for design, in that 80% of a site design can be achieved in 20% of an entire design project timeline. The fine details would consume the remaining 80% of the time, however there are a limited number of people who care about (and are able to) perceive the 20% increase in detail.

This 80% is the theoretical Threshold of Appreciation. It is entirely possible for a designer to sweat out 8 hours when the client would well have been happy with seeing the results of (and paying for) 2. Of course, we’re taking for granted that you’d have to be a halfway-decent designer to be able to knock out something efficiently in a short time.

And good design isn’t difficult to do, really. Good designers understand conventions and use them well; they also tend to have a catalogue of templates in their heads, ready to be dropped onto the screen.

Back in the years when my job requirements (and workload) were more design-focused, getting something out that looked amazing within a matter of hours was a challenge I enjoyed. Clients with large design budgets were few and far in between, and the opportunities to tinker with in-house interfaces were even fewer.

So where I adhered to the Threshold for the Alpha, designing the Beta was initially a fraught affair. It was a little more difficult being pixel perfect than I’d anticipated, and getting through it took me nearly three days of isolation and thoughts such as:

  • I’m spending way too much time on this.
  • What if I don’t have the ideas to be able to come up with the detail required? O god.
  • I feel fidgety; am I spending too much time on this?
  • Nothing’s coming! I’m a fraud! *bawls*
  • Okay this bit’s good enough. No it’s not. 
  • Yes it is.
  • No it’s not.
  • I’m thinking too much and not pushing enough pixels! Push!
  • This feels completely indulgent so it must be wrong.
Anxious, much? Absolutely.

But I think I’m getting better now, and am understanding that putting that little bit extra in is now part of my job description, and not something that I have to sneak into a project whilst sitting in the office in the early hours of the morning or on weekends (so he says, sitting here at 1.56am on Sunday morning – hah! But hey, it’s our baby after all). Moreover we should view the extra as mandatory, not optional.

There’s a distinct difference between designing a simple, consumer-friendly site or store versus designing an application. Again, conventions come into play: most web users are familiar with common tasks such as shopping online and navigating a content-heavy website. Unlike websites and e-commerce stores, our site is our product. When it comes to proposing an opinionated system for managing finances, it’s imperative that our users get the idea as quickly as possible – otherwise we’re going to find acquiring customers pretty difficult. 

Some time back I attended a cooking class (my one and only to date) – and while I have fond recollections of making hand-made pasta and preparing a tom-yum soup entree, the lesson I took away from the experience was a slice of wisdom from the chef.

“A gourmet meal,” she said while whisking some eggs, “must have three components. First, it must smell divine – this is your diners’ first experience. Then, it must look stunning. These two factors combined will whet the appetite. And finally, it must taste heavenly. This will bring your diners back for more.”

I believe these principles are beautifully analogous to what we’re doing. Much like a good meal, the three factors that will keep a user coming back to PocketSmith are all related to the experience: it must look good, work well, and give the user some information of substance. And design is the pivot point of our user experience. 

So PocketSmith’s design is highly important on two fronts. Firstly, the user interface has to be great. Not just good, but great. We want our users to enjoy working with the app, and nothing lets a great web application down more than a poor user experience. The interface is a big part of this, and the best bits lie in the details. The challenge for us here is to convey meaningful information in a succinct but powerful manner, and to simplify processes as much as possible.

A source of inspiration for me during this process has been Edward Tufte’s Visual Display of Quantitative Information (I was given an autographed copy of this two years ago by someone who evidently had some remarkable foresight). Check out Tufte’s thoughts on the iPhone (1.x) in this video; I wonder if I’m the only one who thinks he sounds like Steven Wright.

Secondly, we’re working to establish a strong and unique brand appeal, which makes the task more challenging for us. Adopting the status quo for application design is the path of least resistance – and sure, while the risk and associated design time is low, your product will look like every other one that’s out there. Remember how designers have conventions? Well we’re shying away from the common look and feel of web 2.0 applications as well as one of its relevant subsets, web-based financial planning applications.

This one’s a little more difficult, and the brief has been a little tough for me. It’s way past my bedtime but I’m on a good rant here, so don’t feel sorry for me – I feel sorry for you! And will consequently touch on the brand.

The name ‘PocketSmith’ was selected for a number of key reasons. It comprises two common English words, making it easy to pronounce, remember and spell (imagine trying to market a product called “blargoo”, or “zoowee”, or… some relatively meaningless web 2.0 name).

It’s also a more elegant way to talk about one’s finances – for example, to be “hit in the pocket” is a turn of phrase that describes an expensive endeavour; being “out of pocket” sounds more elegant than say, “I’m f-ing broke”; a ‘pocketbook’ is a purse or a wallet. A smith is a craftsperson. So, a ‘PocketSmith’ is a craftsperson adept in helping one manage their finances.

Given the old-worldly nature of the name, the design brief practically wrote itself. We would attempt to build a financial planning application that hints towards the 20’s and 30’s; sepia tones, art deco, jazz. Maybe steampunk, but not overdone. Something that will visually stimulate our users, but still remain relevant, not date (too quickly) and most importantly, still deliver the goods. All the while realising the art deco has limited appeal.

And yes, James and I have played Bioshock, and while it’s a stunning sensory experience, be thankful we weren’t complete in applying inspiration from this source as it would have without a doubt, freaked our users out. I have yet to read Ayn Rand, but perhaps someday soon. Even so, perhaps a tad ostentatious to claim her as a source of inspiration.

So there it is! Maybe I should have crowdsourced this one, lol.

All in all, the design for Beta 1 is looking pretty good, and is a cut above (high above) the Alpha. There’s been a bit of a push and pull between ’sensible’ and ‘toeing the line’; after all, when an application is somewhat themed it doesn’t take much for it to become a visual cheese-fest. And I’ll trust our Beta respondents to be honest about their impressions.

So what, actually, do pixel-perfect designers do? Well they are anything like myself, we spend hours staring at the screen, poring over the layout as it slowly grows. Just staring, and not so much pushing pixels. Because a lot of the work, it seems, is done in the head – and this requires a lot of hard looking. Also, as evidenced above, there are internal battles to be fought.

3.34am. Goodnight! And don’t forget, kids –  Time Machine rocks :-)

 

Five things to know and do when you’re the co-founder of a web startup while not being a developer nor a designer

Saturday, August 16th, 2008 by Francois

Hey folks,

I have been away for a while now. I have been busy trying to nail my thesis for a week and half and I have done 25 pages so far…(out of 85 minimum ahah ;) I am not going to reveal when is the deadline because I am already too stressed out!

Instead I am going to discuss some real-world issues in this post ( I have been a bit cynical with academic bulls**t recently ;) Something that come back to my mind very often is the fact that I am part of web startup company without knowing anything about coding nor design… Well, being in this situation does not look very comfortable at first glance. EXCEPT if you follow these simple lessons I got to learn in my first 6 weeks with PocketSmith:

  1. GOOD TIMING: (Good) Developers are always busy. Their brains are constantly focus on how they are going to fix this issue that prevent them to sleep. And more importantly, they don’t like to be interrupted when they are trying to fix something, especially if you are not a coder yourself (and French ;) Consequently, make sure you spot the perfect time to speak with them. From my experience, the best time is just after they solve a problem (let them a minute to celebrate and then take advantage of their good mood!)
  2. GAP FILLING: Coders love to spend hours fixing bugs because they get to see their “baby” finally working ( = big satisfaction). Especially if what you are working on is as sexy as what Jas and James are doing! The more they dedicate their time on “hard-coding”, the less they will want to take care of other parts of the business. As the third-man, you role will then be to fill the gap this situation creates. For example, take care of the relationships with your bank, your accountant and your lawyer. Use PocketSmith to record money transactions for the company (useful to claim the GST back at the end of the year) Or set up a testing protocol to ensure the reliability of the solution. To sum up, take the ownership of missions your partners are to busy to do.
  3. ABILITY TO WORK BY YOURSELF: Because your partners are going to be busy, you need to be able to work by yourself. This exercise is not that simple as nobody is going to tell you what you have to do. Consequently, start by identifying what would be useful for the team to have. It could be a competitive analysis, a mail to alpha testers, a draft of the questionnaire you intend to submit or a simple graph with the major milestones of the project. Share and decide what you think is relevant with your partners and do it. Then share your results with the team!
  4. OPEN MIND: When Jas and James speak together, it takes a lot of efforts to actually get what are they talking about. First, they speak in english. Then, they use technical words that sounds like chinese to me! But you have to force yourself to understand the global meaning at least. Listen to them when you have a chance and do not be afraid to ask questions (not to much because coders dont have time :) . Pick up some ideas, try to use the same vocabulary and level your knowledge up. I generally dont participate to discussion between Jas and James when they are coding BUT I do want to share my opinion about the useability of the product. If you want to offer solutions to be implemented, you need to have an idea (even roughly) of the feasability of this solution. Theoretically, everything I can think of is possible to code. However, our company has limited resources (time and people) so we have to voluntarily limit what we build. Having an idea of the complexity of a solution is in this case very useful!
  5. USER VIEW: Coders and designers love their work. They build it from scratch and spend 12 hours a day making it perfect. But perfect does not mean perfect from a user point of view. By not being involved into the development process, I am an (almost) impartial user. Consequently, I can comment on the usebility of the product and make recommendations to improve the user experience. Happy happy user :)

Being honest, it is not easy everyday not being able to help the guys coding or designing. I would like to be part of the development process as it is so rewarding to see the result on the screen. Sometimes, I feel I am totally out of the conversation and isolated from the “core” of the company. But this position allows me to stand back and think about the big picture. And at the end of the day, the three of us have found a really good balance that make PocketSmith already successful. This unusual combination is our strengh. In PocketSmith we trust!

The slide to destitution

Tuesday, August 12th, 2008 by Jason

I’ve been working hard on the data storage, management and interpretation of a user’s actual transactions that are imported from bank statements, and it’s taking quite a bit longer than I had anticipated. The initial upload went pretty smoothly, however the dissemination of said data is quite time-consuming. I actually opened up this post to upload a picture, but there are a number of discoveries that I’ll quickly drop in here to help anyone headed in this direction:

  1. I’m speeding up bulk imports in Rails using FasterCSV (I’m using it for the CSV exports too) and ar-extensions. The former makes importing and exporting a cinch by helping me easily parse rows, and the latter executes some SQL-fu that consolidates inserts, making bulk updates super-fast when compared to the traditional Rails method of looping and saving. I doubt I’ve scratched the surface of what ar-extensions can offer, but I’m very impressed so far as we’re dealing with a lot of records due to the granular nature of our cashflow forecasts.
  2. We’ve been having issues with the sqlite3 + Rails installation on Leopard, with PocketSmith throwing odd 500-code “SQLite3::SQLException: SQL logic error or missing database” errors, which disappared upon a refresh. The errors persisted for me even after James upgraded us to MySQL (which was a good move), and the more detailed errors allowed me to track it to the lack of having native C Ruby-MySQL bindings. The fix is nicely documented on Anil Kandangath’s blog.
There. Now, what does “the slide” have to do with this post? 

Well, I’ve finally shaped the data well enough to pipe it to a rough dual line-graph to compare actuals to forecasts, and the first rough-cut of the graph showed me this (click on the image for an enlargement):

How low can you go?

How low can you go?

Yes, dear reader, those really are my actuals. Albeit mapped against a fictional forecast and starting balance. However, any donations of catfood and hugs are still greatly appreciated.

I admit I was stupid, but now I am back up to full speed

Tuesday, August 12th, 2008 by James

So about three weeks ago now, I spilled beer on my MacBook Pro (which was at the time less than a month old) on a Friday. Don’t worry about berating me; I did enough of that in the hours that followed. Luckily there was no great *poof* of the magic smoke escaping and I turned everything off, mopped up as much as I could, and let it sit to dry overnight.

Next day everything started up perfectly, however some liquid had gotten in the LCD of my beloved laptop; while this dried up over the next few days there was still a decent amount of residue that was extremely frustrating. So I called the good people who provide my household insurance, and luckily enough this level of amazing stupidity is covered under my policy. I think I even paid for it in my premium, I believe it is a policy addition called “Clumsy Fool Cover”. Everything packed up and sent away to them.

In the interim I reverted to my trusty 1.42GHz iBook, which meant that I could continue working and coding on PocketSmith. Now I do say I could continue working, however I wasn’t able to be as productive as I once was; I had very quickly forgotten how painfully slow this pretty-up-there laptop from 2005 could be. Solid as a rock, but some activities take forever.

And being in heavy development mode at the moment, there is a lot of restarting local servers, refreshing pages, and heavy calculations being performed by the local server on the iBook. Whereas previously a quick change of the application’s environment, a restart of the local server, and page refresh would have taken around ten seconds on the new shiny MacBook Pro; the process would take around one and a half minutes on the iBook.

The machine was so slow that at the end of the time with the iBook I had taken to actually pushing changes I made locally up to the live server using Git to test changes, because it was actually faster to do things in this round-about away!

Another example; we are using the (fantastic) services of Y-Combinator startup company Heroku for the hosting of the application in these early stages (and hopefully the later stages as well; more on this later). They have this pretty nifty in-browser editor for rails projects you have in Heroku. And I thought it was broken; loading the editor would take 45 – 60 seconds each page refresh, with the page slowly coming together and being functional in about the time it takes to make a cup of tea. Well, a very quick and messy cup of tea, but a cup of tea nonetheless. I had initially thought there was something wrong with the way that Firefox was caching the Javascript. How wrong I was.

It was simply the length of time that it took for the iBook to use the Javascript to render the page; once I got my MacBook Pro back (new top-shell, screen and left hand side speaker, hoorah!) I realised how blazing-fast the interface was, like 15 times faster.

Funny thing is that at the outset I was wondering if I should bother with a Macbook Pro or whether we could start this thing with myself on the iBook. So now I know – if I had taken that decision, we would still be in pre-alpha instead of rushing headfirst into the beta :)

Reconfiguration for the beta

Friday, August 8th, 2008 by Jason

After an amazing indian summer day yesterday, it’s cold and rainy again out there as I sit in the office mulling over our upcoming csv import feature. Francois is (hopefully) taking the day off to work on his thesis, and James is getting some rest after pulling another marathon night cooking up some UX magic. 

I’ve taken to pulling down the latest development version of our app in the mornings and seeing the fruits of those late-night endeavours with a grin on my face. It’s like Christmas morning all over again!

Anyway, this is just a short post to let you know we’re still here.

There’s been a bit of radio silence while we’ve been putting our time into building up the features for PocketSmith Beta 1, due September 1. We’re getting increasingly comfortable with Rails, and are realising day-by-day how powerful its conventions are. There’s still a lot to learn but we’re enjoying our journey. We’ll post soon about the development environment that’s helping us bring PocketSmith to fruition.

Yesterday the three of us spent the afternoon in the sun on Francois’ deck at St Clair thinking about how to structure the beta. The alpha was reasonably straightforward, bearing three key features – cashflow input, forecasting and graphing. More of a technology and usability test than anything else – although, we had the audacity to ask our participants in our questionnaire if they would pay a monthly subscription for it!

We also had fun with the user interface (a quasi art-deco/steampunk approach), but now that we’ve got to hinge a lot more features onto the app, the interface needs to be flexible enough to allow components to snap in and out of it. By Monday we’ll have half-a-dozen new features lying around that will need to be coherent and make sense to the user.

So it looks like we’ll be dismantling the application, redesigning the UI, then… remantling it. Remantling should be a word. I think it works.

I hear my arrays calling me.

The Techcrunch interview

Monday, August 4th, 2008 by Jason

We’ve been laying low for a couple of days since the alpha launch, taking it easy and thinking about the next steps in PocketSmith’s progression. The feedback from our great alpha testers is rolling in, and from the initial comments alone we’re already taking in some valuable tips on our future direction!

More on that at the end of the week, when we can take stock of all the data. We haven’t yet told you how the TC50 Round 1 interview went!

So, Saturday morning, 10am(-ish), I went round to the office and woke the boys up – we’d all had a pretty long night (long week actually), so coffee was the first thing on the agenda. When we were sufficiently awake, we thought a bit about how we were going to handle the interview. We’d loosely figured that our 10-minute interview should consist of a 4-minute video, 2-minute chat and 4-minute live walkthrough of the application. Kinda.

Yes, absolutely, we were less-prepared than we should have been. Roll back to yesterday.

The Day Before

After we prepped the application, sorted out the last-minute bugs (see post below) and sent out the alpha launch email, we immediately set to task creating a screencast movie of our application. A screencast generally consists of a rehearsed (and preferably non-erratic) sequence of mouse movements and clicks across the application’s interface that is recorded by an application – in our case, the very capable Screenflow. This capture is then matched up with a voice-over narration.

So we scripted the thing, then made a couple of attempts at talking the user through a 4 minute 22 second video – with hilarious consequences. We were tired and delirious; there was screaming and hooting; James – who has a great radio voice by the way – made it to just after two minutes before degrading into something that sounded like blarghblahblargh. Blargh. Blaaarrrghhh. Followed by ape-like hoots in the background (that was us). Not good for a pitch.

Then James took a break, and I decided to give it a go. You won’t even want to know what I sound like by comparision. I briefly contemplated putting Francois on the mike, but then thought the better of having the evaluators swooning over his delicious Pepe Le Pew accent (”Le mew? Le purrrr! Mwa mwa mwa!”). 

With energy (and sobriety) running low, I prodded James and said, “Let’s give this thing the DVD commentary treatment” – which we did, and we think it works! There’s laughter, teasing and involvement from all three of us, and is much more entertaining than a staid 4-minute voiceover of the mouse-clicker’s actions.

At that point, completely exhausted, hi-5’s all around, we decided to call it a night.

Cue Saturday morning. An hour before the interview (scheduled for 11.45am), we were still looking up web-based powerpoint equivalents to beef up the presentation; 45 minutes before the interview, we decided to can the search and just ‘be ourselves’ (the de-facto strategy of many a last-minute presentation). After all, we would just be chatting to a couple of Techcrunch evaluators – no big deal, right?

I was in the kitchen pouring us coffee at 10.30am when my phone rang. I ran back into the office, picked up the phone and put it on speaker. On the line was Jason Krute, calling ahead to to ensure that our interview materials were set up. We opted for Skype as the connection seemed pretty poor – this turned out to be a great choice as the connection was crystal-clear, and all three of us could partake in conversation. So there we were, grinning confidently, ready to have some fun – until Jason says, “Okay, I’m going to put you on mute until your interview begins. Your evaluator is Jason Calacanis.”

Brix. Were. Shat.

Our interview with Calacanis

It was a giddy three minutes while we waited, pondering what the interview was going to be like. To be honest, I for one thought that we were going to get slaughtered. The man sees a ton and a half of startups on a daily basis, and is well-known for being opinionated. He famously stated, for example, that SEO was bullshit – in a keynote at an SEO conference with Danny Sullivan.

Our Skype conversation clicked back on, and Jason transferred us to Jason (yes, too many Jasons in one conversation), and the Mahalo logo popped up on the interface.

Jason Calacanis sounds much younger than you’d expect. And he’s terribly polite, too. He greeted us briefly, then told us that he had the link to our video, and was going to watch it. So – the three of us spent another four minutes on mute – partially thankful that we had our tequila-addled representatives from the night before doing our presenting for us, and partially cringing that perhaps, the tequila-addled representatives weren’t taking the TC50 application seriously enough.

Skype clicked back on, and Jason said, “So let me get this straight – PocketSmith is about expense tracking, am I right?”. And I corrected him: “Actually,” I said, “PocketSmith is about cashflow forecasting – it makes financial management accessible to people who are not necessarily financially literate.”

Then I went on to quickly run through the key points we’d noted half an hour earlier – the purpose of the product, how it was meant to achieve our goals, our target market and… “Yes, ok, I’ve got it”, he interrupted. “You don’t have to read me your entire business case, we’ve only got a few minutes and I want to give you some feedback.”

“It’s cashflow forecasting that doesn’t suck, right?” he said, and we quickly agreed. He then ran through a few of his notes summarising what he thought our application was about, which we thought was spot-on. Then, he gave us our feedback, which I’ll attempt to summarise.

“Your demo was okay,” he said, “but not amazing. It looks good, and you’ve made a good start, and I think you’re in the right space. But you need to have that wow factor for the audience. The sizzle.”

He went on to stress that while we had made a good start and that he enjoyed the video (!), we needed more innovation in order to increase our chances of making it to the final 50, and provided some suggestions: syncing with his iPhone and calendars; import of spending reports from banks; tips on savings. He then asked if we’d considered any of the above, to which we replied that the application had only been five weeks in development – however we did want to differentiate ourselves from the likes of Mint, who already do imports and financial advice very well.

He agreed, stating that the concept was simple, which was not a bad thing. And reiterated that Mint did after all win at last year’s Techcrunch conference, and that he thought that we were pretty much 10% of what Mint had to offer.

Our five minute chat with Jason Calacanis felt like much longer, and he made a concerted effort to keep us on-track and on-time (they have to be strict on the 10-minute timeframe), and he ended by telling us that while he couldn’t promise that we would see it through to Round 2, if we did, we should look into innovating more on the product. He finished up by telling us again that we’d made a very good start, and thought that PocketSmith would have been great for him back in ‘95-’97, when he really needed to manage his finances! 

Through the interview I was very tempted to drop a hint about how we’d heeded his ’saving money’ advice from his “17 Good Tips for Startups” post (given we were conducting the interview from a makeshift fibreboard table that James put together for me), but realised that it wasn’t advantageous for me to waste our precious time with adulation. Besides, my butt was firmly seated in a $45 plastic swivel chair, and not a $1200 Herman Miller Aeron. Which in hindsight, may have won me some commendation – damnit!

After thanking us (many times) for taking the time to submit an application (no, thank you!), we bid Jason Calacanis goodbye and hung up. Then stared wide-eyed at each other.

Summary

What can we say? We completely agree with Calacanis, and felt that his insights were – as with his initial summary – spot on. His assessment of our alpha is completely on-track with our development to date, his suggestions were on our list of considerations, and most importantly: he didn’t think that it sucked. While we wait to see if we’re pulled into Round 2, we’ll need to work on that innovation bit. 

We also came to realise that while Techcrunch is a great platform for launching a product, the fact remains that it still is a conference – and presentations at conferences tend to be noteworthy, like new and important discoveries, and things that generally push boundaries and challenge perceptions. PocketSmith doesn’t necessarily fall into these categories as we’re not looking to debut something with a wow factor for an audience. We’re committed to building a great product with high utility value – and these factors alone don’t necessarily translate into very exciting presentations!

Anyway, five weeks in, and we’ve built an alpha, released it to a close community of testers, and had direct feedback from one of Silicon Valley’s prominent tech entrepreneurs! Neat stuff!

What now from here? Well, we get back to work! The three of us have decided on the next little iteration that we’re going to add to PocketSmith while the feedback from the alpha continues to roll in. It’s going to be a cool one, and we’re looking forward to building it.

10 lessons learnt from the Alpha release…

Saturday, August 2nd, 2008 by Francois

The Pocketsmith Alpha is finally live! We have been up almost all night getting this first version ready and we are excited to present early Pocketsmith-enthusiasts the application for testing! It is now time to stand back and look at what we have done good and more importantly what we should improve in the future.

5 things we could and should have done better:

  1. Did not evaluate properly the workload left. I still remember us a week ago being sure we had heaps of time to get it done. Yeah. Right.
  2. Did not respect the planning we were supposed to follow. We did decide together we will have a least a buffer day after testing. We ended up still testing the app 2 minutes before sending the email to our alpha testers.
  3. Did not set up a testing protocol. That is such a newbie mistake ahah. But I guess this is what you should expect with a Frenchman in charge of this job!
  4. Adding features at the last minute. I swear we tried not to! Unfortunately having high expectations for this version means we wanted to make it as useful as possible.
  5. Assuming that starting on the first day of August would not have a major impact. That was bad. I mean bad bad. 10 minutes after the first notice to the alpha testers, I spotted a serious bug related to the date. Yes, we fixed it within twenty minutes but stress levels were pretty high for that period!

5 things we managed well:

  1. Set up our expectations very high from the beginning. Targetin high  helps us to push the boundaries. We have reached this step in 5 weeks. Let’s see what we can do in 10!
  2. Release an Alpha version early with limited number of trusted testers. This step is important as we gather crucial feedback early on the track before facing the “big test” of a public alpha release.
  3. Select a broad pool of alpha testers from various age, gender, occupation and localization in the world. That will give us enough marketing materials to play with :)
  4. Survive the intense pressure associated with this release. We had some pretty intense moment of stress following nasty bugs and unexpected delays.
  5. React fast to the running balance issue. James and Jas have been impressively quick finding a fix. We also took the time to inform the testers about the evolution of this issue. We love our users!

We have had 3 very intense days and I think it is now time to get some rest. This week end will be perfect to stand back and come back on Monday with awesome new ideas!

PocketSmith: The first pictures!

Friday, August 1st, 2008 by Francois

Feeling guilty of not being able to help Jas and James cleaning up the code, I have decided to upload these picture from Jase’s iPhone! It’s not a full photo album but I reckon it is a good start :) We will had new ones in the coming weeks so stay tuned!

Francois using mac gears!

Francois in Night Mode!

James the dark spongebob!

James the dark spongebob

Flashy mac!

Jason's nightclub