Question: TDL Web Version

[JW.]

Hi there,

After a period of time using the software, I realized that it's a very useful tool and more important, used by many people all over the world. TDL has a very strong community with active contributors. So what I'm about to ask is: Is there anyone interested in developing a web version of TDL? So in the future, it will be possible for TDL users to storage their tasklists, time tracking logs, notes... etc... on the cloud?

Please share your thought!
Best regards,
Alex

[.dan.g.]

Hi Alex

Unfortunately designing for the web is wholly outside my skillset, and that's a new trick that this old dog ain't gonna learn!

However I see no reason why it could not be done, though of course it would probably not be free since the app itself would have to be hosted somewhere.

Dan

[JW.]

This is a discussion, not a request so you don't have to worry about doing it :D

The fact is, I had recently seen a few GTD websites. Of course, the software is good in its manner but having something on the fly is convenient too.

[Tony G]

Alex, one of the reasons I created ToDoListLib .NET (not yet production, sorry) was to facilitate the development of alternative UIs by other developers. With the library as a proxy between a website, mobile app, web services, or other clients, TDL gets a whole new life outside of the desktop.

Another angle on that - I've recently subscribed to Todoist.com to see how it compares to TDL. They also have a developer API. One of my goals when I get more familiar with that service is to create an interface between TDL and "TDI" so that we might be able to use their UI for our data - with some limitations on their side of course. ;)  That provides a new Web UI, cross OS and cross-device mobile apps, and a number of other interesting features without the burden of doing it all from scratch. The trade off is that someone using this would have no control over that UI and their premium service is US$29/year.

In the big picture, I'm hoping that when the Library is firmly established that a single code set might serve as the rule base for a number of new UIs. So for example, someone might create a website, and someone else might create a new Android app (there's already one out there but not fully compliant with the latest .tdl schema), but both might connect to a server that provides a common set of rules. That leaves developers to writing a pure UI using POCOs and POJOs to interface via web service - very MVC-style. If the Library gets changed, no changes should be required to apps that use it unless there is a breaking change to the data model.

Final note: I mentioned there is already a TDL Android app by a third-party developer. One of my thoughts has been that if a single FOSS HTML5 GUI is created, it might be ported into apps that run natively on devices and websites. The idea is that while some developers are eager to create their own special view of a UI over TDL, most users just want to access their TDL data and they're not looking for different apps for different platforms or some creative angle on the model. So, like Todoist.com it would be nice to get a consistent UI across all platforms with a single codeset (as close as we can get that). When a change is made to the UI core, developers can port it into their platform-specific implementations. Then, if developers want to go off and create something unique, great, but at least we could have this as a baseline implementation, with support for a wide variety of UIs, to set the example for others to follow if they still feel a need.

That's how I see some of this playing out in an ideal world. Lack of funding and time means all of this remains little more than a collection of ideas for now.

Thoughts?
T

[Joey]

Hi Tony,

Actually todoist.com is a very neat way to handle GTD matters and keep your personal list on many different places. Also - I guess - well worth the money you need to spend for full usage. The todoist.com UI is also quite good from my point of view.

However, the way I see it, ToDoList is more powerful and far more customizable.
(I still have to delve into it, to check if there is a proper way to edit the UI, though).

I just recently decided to use ToDoList, after I tried a number of alternatives; todoist.com among them.
For me the two main reasons for choosing ToDoList were: private data, stored in an open format and customizablity.

Call me paranoid, but I like to keep my data (especially GTD) private. Now way of storing my client's matters on some else's cloud. However I would love to be able to update my GTD list online, using a smartphone or tablet. Thus a web-interface would be great, and I absolutely second that idea. Making it instable on standard webservers, the users could choose a hosted service (they thing they can trust) for a few bucks, or setting it up on their own server and keeping control.

Lack of funding is an issue. Like always, yes.
But still, I would like to edit my GTD list online and will further think about this.

Cheers
joey

[Tony G]

There are various options depending on developer skillsets.

If you work with Java then perhaps you would consider offering the developer of the Android app assistance. He was working on his app as a personal pet project but I think he got turned off when people started asking for changes - as happens to a Lot of FOSS developers. Unfortunately that app has drawn some criticism because it isn't fully compatible with Dan's ToDoList, and I fear the app functionality may reflect upon Dan's work even though he's not related to it in any way.

You could dabble with the TDL source but I don't believe the UI is modularized far enough away from the rules (MVC-style) to allow you to abstract your own version of the UI without significant retrofitting on every update from Dan. It's just not worth it. If you could plug Dan's rules into your UI on every update, you'd be golden, but that won't happen.

I believe Todoist has an API (one of the reasons I signed up with them too) but I haven't looked in a while. Perhaps you could work out XSLT to transform between the .tdl schema and theirs? (Or any other) That would be a valuable contribution and would allow you (for example) to make changes in TDL, save, auto-export to Todoist, use your data via remote/web, import back to XML locally, transform, then pull it back into TDL.

Another weird option would be to use something like WordPress or another CMS. Create custom forms and fields to mimic those of TDL and publish them on a private login. The data can be hosted anywhere you wish, shared server, private, whatever, it's all separate from the website. And again with import/export between MySQL and XML the data can be transformed to/from a .tdl. This could be done on each record update, triggering a transfer of the XML into a dropbox where it could be opened by TDL, with no special transformation effort at all. I could do this, if someone put a gun to my head, or enough beer and pizza in my mouth. The UI would be icky (um, comparable to the Android app?) compared to TDL but it may be better than nothing.

My vision for the Library is to do that transparently using OOP with established APIs rather than a more clumsy transformation approach. If some company offered to fund development for their usage, we'd all be Very happy. Unless that happens we all need to wait for "free" time where I can get this finished and published as a v1. Dove-tailing with the above, the Library is a quick vector to interfacing with a database (SQL Server, MySQL, etc), providing the transformation layer described for WordPress (Joomla, Mambo, Todoist, whatever).

So is your level of skill in various areas? Perhaps we can find some point where you can help to create this beast that you'd like to see.

T