Guerrilla computing in the mountains of Seattle
The Mentat web service has been updated to include contexts, subprojects, and data export.
Coinciding with this, Black Pixel Luminance has announced Aria Touch, a native iPhone application that supports both Mentat and FogBugz.
Filed under: mentat by Daniel
No Comments »
The past couple of days I’ve been working on adding “contexts” to Mentat. This key component to the GTD methodology was a major feature request so it was first on our list of things to add to the Mentat system.
Contexts in Mentat will work like this - Although you have separate projects to keep all of your varied tasks separate and distinct (Work Projects, Grocery Lists, Chores at Home) - there are contexts which might be project independent and could help you filter down those tasks you can take care of when you are in a certain “mode”.
For instance, say you have some time to make a couple of phone calls - and luckily enough, you’ve made a special context called just that “Phone Calls To Make”. By pulling up the context view in Mentat, you will be able to see all of your tasks that you’ve marked with this context (whether they be phone calls you have to make for your “Work” project or your “Home Chores” project). Now that you are in a perfect opportunity to deal with this “context”, you can fire off all these tasks at once!
Contexts can be assigned by navigating to a specific task view. Or, in the Mentat Agenda, you can quickly switch between a project and context view to quickly drag tasks into your agenda using either filter.
I’m almost done incorporating this functionality - just doing some polish now. I hope to have something by the middle of next week, so stay tuned!
Garrett
Filed under: mentat by garrett
No Comments »
We’re moving back towards a focus on updating the Mentat website now that we’ve got a few mobile/desktop Mentat-empowered clients under our belt (Mentat Blackberry, Aria, Arioso). Today we compiled a list of feature requests and began prioritizing what we felt was most dire in building out and improving upon. This is what we came up with:
1. Contexts - As a GTD cornerstone, this looks to be a number one priority for our customers. We’re working on this one first and already have some functionality prototyped. With this feature, you will be able to lump tasks across projects into specific “contexts” such as “emails to write” or “phone calls” so that you can blast off otherwise unrelated tasks all at once.
2. Third Party Agenda Tasks - With the introduction of Aria, Mac OS X users can have access to both their Mentat and Fogbugz tasks. We’ll be building in support for generic third-party tasks and adding the ability to include them in your Mentat agenda.
3. Subprojects - Since the dawn of Mentat, we’ve been wanting to include this. Now is the time.
4. Email to Mentat - We’ll be adding the ability for users to use their email to create and respond to tasks in Mentat.
We’re still taking product suggestions, so feel free to drop by the forum and let us know what you’d like to see next in Mentat.
Filed under: mentat by garrett
No Comments »
We have been spending some time working on general performance improvements in Mentat. You should be able to see the fruits of our labors immediately.
Both the Agenda page and the Projects page have received some love. The loading times have been drastically improved, and we should no longer be getting in our own way when it comes to element caching.
We (re)discovered several things —
And now back to your previously scheduled program. If you didn’t have a previously scheduled program, allow us to recommend Battlestar Galactica. It, like Safari, kicks butt.
Filed under: Dave, mentat by ddb
No Comments »
There has been a lot of bitching about the terms under which developers can sell their iPhone applications through the App Store. I would like to share my opinions on the matter, coming from the perspective of a developer with experience bringing multiple commercial products to the handheld market.
Sales channels make their living by taking an existing application and getting it in front of a vastly larger number of people than your company can. This isn’t just important to you: if you’re looking for an investor, they are going to know what sales channels you are going through. It’s the difference between a modest success and a blockbuster. If the sales channel does its job well you can hit a much larger market than you would ever get on your own.
Let’s take it as a given that you have a great product: no one can buy it if they haven’t heard about it. Sales channels can get the word out on your behalf and you don’t have to lift a finger. You can focus on your own marketing plans, or keep writing your code or whatever.
One of the biggest sales channels for handheld software is Handango, which takes a 40% cut out of all of your sales revenue, not including taxes.
This is not limited to handheld development, either. If you have seen Wil Shipley’s C4[1] presentation, he discusses the cost of working through sales channels, and why they are such a big deal, starting at around 1:04:40. The groups he worked with came out to around 50% just get your products to the sales outlets. The entire section is really worth watching if you are serious about getting your product out on the market.
It gets even worse when you enter the field of physical distribution of goods. I have heard horror stories about the margins demanded by stores that stock items on shelves.
But enough hearsay — I have a friend that runs a specialized book publishing house, and when he sells books to dealers and stores, he receives 38%. That’s a 62% margin that the channel is taking.
We have it pretty good, all things considered.
Great Exposure
Apple is willing to feature your application at the App Store. That’s right - someone is going to buy an iPhone and then start looking for some apps to put on that baby, and *your* app will be listed in their database.
If you have ever tried to get your application featured on a RIM- or Sony Ericsson-hosted ‘applications for your new phone’ site, you will realize what a huge, huge deal this is. Traditional companies require a vast amount of QA, frequently an expensive membership, and occassional, a sample of your brain matter, in order to get your application hosted on such a high profile and easy to find sales outlet.
Obviously, you’re going to be surrounded by a ton of other apps competing with yours, but that isn’t Apple’s problem, it’s yours. You are still listed at the first site customers are going to look to for new software and that’s golden.
Reasonable Prices
The App Store will be taking a 30% cut of sales revenue on commercial iPhone applications. I’ve been hearing a lot of people grouse about this, and, on the other side of the fence, a lot of pro-Microsoft people jeering about this.
Let me set the record straight: as a sales channel, Apple has picked a very reasonable price (particularly considering that they charge nothing for handling free applications). Those that complain about a 30% commission are merely revealing that they haven’t tried to bring a serious commercial application to a large market themselves.
Sure, Apple’s Macintosh Products Guide can list your Cocoa application for free, but they aren’t hosting anything or brokering anything, they’re just providing a blurb about your product and a link to your site. All of the ecommerce headache still has to be handled somewhere, by somebody. Probably you.
Most of the people that I’ve heard complain about the 30% cut haven’t actually brought a handheld product to market. I agree that it might be nice to be able to sell products on other sites, but the only site I’d really be interested in doing that on is my own. Anywhere else is going to cost *at least* as much and wouldn’t get anywhere near the exposure or easy customer access that my applications will get at the App Store.
-Daniel
Daniel Pasco is the CEO and chief engineer of Brain Murmurs, Incorporated and has worked in handheld software development since 2004. During that time he has developed three commercial software products for the BlackBerry and one for the Sony Ericsson P900 series of smart phones. He is currently working on a iPhone task management application based on the Mentat web service.
He has still not gotten his code signing certificate from Apple and is concerned that it may miss him.
Filed under: Dan by Daniel
No Comments »
Development on iMentat is going *fast*. It is very likely we’ll be ready for beta testers by the end of next week, if there’s actually a way for us to distribute the builds to people.
*Screenshots have been removed on the advice of a colleague until we get some clarification from Apple on whether or not we can post them.*
Technorati Tags:
GTD, iphone development, iphone sdk, Mentat, organizer, user interface design
Filed under: mentat by Daniel
No Comments »
The Mentat BlackBerry OTA download links are working again. Turned out that RIM had it set up to secretly split the extremely important .cod file into pieces once it reaches a certain size and then disguise it after building. After discovering this hidden gem of arcane knowledge, we devised a solution to get things working again.
So the short of it is, technical problems on the RIM side, all is well now. We’ve downloaded from the OTA links onto all our own BlackBerrys and welcome you to join us.
In light of these technical complications, we are extending the price increase for the new build of Mentat BlackBerry to 3/17, so it will go for $25 a while longer if you want to get in early on the upgrade. The latest build is of course free if you already own a registered copy.
Erik
Filed under: mentat by erik
No Comments »
Before we get to the meat of this post, let me say:
Hello! My name is David Brown, and I’ve joined the fine team here at Brain Murmurs.
On to the post.
One of the fun things about my job is that now my interest and my vocation are lined up. Before, whenever Apple released something new, I would have to wait until I got home to play with the new toy. Now, part of my job is to stay on top of these developments, and actively evaluate their impact and usefulness to our business.
Yesterday Apple released an early beta of the iPhone SDK. And what a release it is. There are mountains of documentation, a nice series of introduction videos (which I spent most of last night watching, and I’m only really a third of the way through), and a simulator that you can use with the Xcode debugger.
When you look at the sum total of everything they put together for this release, it’s not all that surprising that it took as long as it did to get it out.
But they did more than just get the SDK out. For anybody that has been following the grand “Hacking of the iPhone” — jailbreaking their iPhone, building the toolchain, and reading the writings of developers like Erica Sadun, the API being presented in the official SDK is quite a bit different than the API that we’ve come to know over the few short months we’ve had an unofficial development environment.
Some of those changes are amazingly great, like UINavigationController and the like. But others, like the requirement for no background execution, the complete sandboxing of the application, the inability to add state icons to the status bar — those changes give us less utility than we had before, when we were exploring the unknown frontier of the device.
I can understand why some of these changes are being made. They make it less likely that you are going to inadvertently kill your device. You cannot destroy the essential data that resides on your device.
Yet I can’t help but feel that we’ve been given the blunt-nose scissors to go with our fine iPhone construction paper. Yeah, there are plenty of very satisfying things we’re going to be able to create. But there’s a whole boatload of applications that are closed to us. Do you want an instant message client that will tell you that you’ve got a message even when it’s not the frontmost application? Sorry, you are out of luck. How about the last.fm client that uploads what you are currently listening to the last.fm service? Not possible under the current model.
Deep down inside, I’m a systems level programmer. I’m most at home with sockets, threads, low-level access to devices, that sort of thing. I love being up to my armpits in memory dumps. While I still have plenty of ideas that I can implement very satisfactorily with the SDK as it currently stands, I ache when I ponder the things that I can’t do.
For the most part, Well Done, Apple. But please, give us a reasonable path to the rest of the device. There is so much that this device can do. Let us prove it.
Filed under: Dave, iphone by ddb
No Comments »
I’ve had about half a day to look at the iPhone SDK and start migrating our iMentat code over to it from the toolchain, and I already have a few observations to make about it.
Working with an actual SDK, and a debugger, with actual documentation, just about made me swoon.
Okay, a lot of stuff has changed. I talked with David Brown last night after working with it for a few hours and it became evident that a lot of APIs have been moved, refined, and in some cases, simplified. A lot of these changes made sense - name spaces have been standardized and are more in keeping with the desktop Frameworks.
Other changes also make sense, but apparently come more out of a desire to impose design standards on 3rd party developers than anything else. For instance, it is now assumed that all tables will have only one column, which was not true with the reverse engineered toolchain APIs, where you could create an add multiple columns pretty easily.
As a result, most of our front end code won’t compile and will have to be re-written from scratch. This is, of course, made a hell of a lot easier by the presence of actual, physically arousing, documentation.
On the other hand, our Mentat OS X framework compiled without any hiccups and appears to work well in the new simulator, so that decision choice was a very big win for us.
While reading through these documents a few things became clear: although we have made a move up from steerage (web apps), we’re still definitely in 3rd class on this boat. Our applications are basically allowed to run with extremely limited privileges and are forcibly killed when the user returns to the home screen. You can change the look of the iPhones status bar but you can’t actually add or update any icons for your own application.
For an application like Mentat, this is an instant condemnation to total mediocrity. Part of what makes Mentat such a great application is its event handling. When someone in one of your shared projects foists a task off on you or leaves a comment cursing you for the task you foisted on them, you get a notification. On the BlackBerry you get two distinct audio alerts for Mentat events; one for new comments and the other to let you know that one of your team mates or your domestic life partner has just dumped a bunch of work on you.
This is made possible by the fact that, on the BlackBerry, Mentat can run in the background and will periodically poll the server for new events. Mentat isn’t alone in this, Twitterfic does the same thing looking for new tweets. Most instant messengers work the same way.
Since we can’t run in the background, the only option users have is to keep our application running in the foreground constantly if they want to be alerted of new events. This is, you may recall, one of the things that sucks about web apps. Which kind of puts us right back where we started.
Another feature that our BlackBerry customers demanded early on - and now enjoy - is the ability to create a task directly from an email or SMS message. At this point it doesn’t look there will be any ability to add a similar feature for our iPhone customers any time soon.
I’ve been developing BlackBerry applications for almost four years now, and Apple has definitely taken a page out of their playbook for dealing with iPhone applications, but they still don’t have feature parity with what a BlackBerry developer can do. When you think about it, this doesn’t really make too much sense, because RIM has always been at least as sensitive to security issues as Apple is being now. RIM’s always had to deal with the same problems, but has still found ways to get developers most of what they need to write great applications.
By default, any application that you write with RIM’s JDE will run on a compatible BlackBerry will run without any code signing. This provides you with just about as much capability as the iPhone SDK lets you have.
You can, however, get much tighter integration with the BlackBerry’s lower level features by ponying up $100 to RIM and getting a code signing certificate. This lets you write applications that can integrate with email, interact with Bluetooth hardware, play around with the status icons, and lots of other fun things.
We know from working with the toolchain that there are APIs that will let you manipulate your own icons in the status bar, and that applications can be run in the background without being terminated. For applications like ours, or Twitterific, we really need the ability to run backgrounded in order to realize our full potential.
One potential way to deal with this is to re-expose some of these APIs to Enterprise applications and handle their behavior at runtime differently. Springboard could examine the application at run time and determine how to handle it - either terminate the application after a polite wait, or simply issue a notification to it that the user has returned to the home page and let it figure out what to do itself.
Joining the Enterprise developer program costs three times as much as the regular development membership but I would do it in a heartbeat if it meant that we could have access to this kind of functionality.
Technorati Tags:
brainmurmurs, Mentat, os x, iphone sdk, apple
GTD, iphone sdk, os x, iphone development
Filed under: Dan, mentat by Daniel
No Comments »
Mentat BlackBerry OTA download links stopped working yesterday and are we are diagnosing it now.
In the meantime, here are desktop download links as an option for anyone who wants to download the latest build now:
*OS 4.2 Desktop Download Link
http://brainmurmurs/products/mentat/v42desktop.zip
*OS 4.1 Desktop Download Link
http://brainmurmurs.com/products/mentat/v41desktop.zip
Just download from that link to your computer then unzip, and point your Desktop Manager application loader Add button to the .alx file and you should see MentatBlackBerry ready for install.
Filed under: mentat by erik
No Comments »