ClikClock has been trundling along with me being able to dedicate approximately 4-5 hours a week to get stuff done.
I built out the account settings code at the end of last week so now we have account configuration working within the site, this will be useful for adding/removing certain options/features on individual accounts as an administrator, as well as the account administrators themselves being able to set certain preferences for their accounts.
Having a single settings mechanism for settings that are read only for the users but editable by the site administrators, or the account administrators presented a few challenges until I hit upon the idea of simply having the ability to set “edit roles” for a particular setting. So I could pull up the details of a setting and be able to say “this setting can only be edited by this role”. I have this implemented and am looking forward to seeing how it works out, there are some obvious permutations to doing something like this but I don’t want to dive too deep on this functionality until I am sure it will be useful.
I have updated my git repository with the latest changes and offered a new dump of my codebase to the other challenger in the 11-05 challenge. I have sent over dumps of my current code before in an effort to help him out in this challenge, but I suspect he takes the code simply to avoid any hurt feelings. HA!
Plan for this week :
Finish the New Account Setup wizard – there are a few features/functionality I think I am going to have to add to the framework to finish this task so I’m going to keep this item as the only todo until my next update.
I was considering ways of generating trees based upon tags and realized that there’s probably two ways to go about generating them. What brought me up short and made me think about writing this down was the following thought : ”when considering the hierarchical implications of tagging what we’re really doing when we tag something is indicating it’s ancestry”. I’m not sure this is really a new thought but it does lead to some interesting thoughts.
- We don’t generally consider tagging to be a hierarchical activity, but what would happen if we just tried some different models for constructing hierarchies from tagging. A perusal through set theory might yield some interesting thoughts on this idea.
- The whole concept of tagging is that we can add an ever increasing number of tags to an item to allow us to describe at some “meta” level what that item means to us. It’s your basic set theory sort of thing, I’ll add an item to this group and this group and this group. From a hierarchical perspective you could say that you are assigning a large number of parents to a particular group, and this seems goofy when you first consider it. This is probably due to connotations that ancestry brings up in our minds. Our DNA is determined solely by our ancestry. Consider a virus though, it can pick DNA up from a bunch of different sources and not just from it’s ancestors. I think I’m digressing.
- My initial idea of implementing a hierarchy was to simply take the tags and show them as children of an item, and to then do some funny business to show things farther down in the hierarchy with some algorithm to avoid duplicates. When I considered the point in item #1 above I realized that perhaps showing ancestry “downstream” instead of “upstream” ( that is, as branches/leafs of the subtree under a particular item ) might have some interesting implications.
I’m not sure there’s anything of value here, but it’s been interesting thinking about it..more later…maybe.
I haven’t been working on the project much the last few weeks, been doing a fair amount of getting out of doors while the weather is nice and soaking up some sunshine and a fair amount of beer.
In the last couple of days I was able to get the migration/deployment stuff figured out for the application which was the last piece that I wanted to have in place before I started pushing alpha builds to a public server. So I anticipate having a publicly accessible demo of the product within the next month or so.
The major pieces I am going to be working on next are :
- Signup wizard : I need to have a way to guide new users into the creation of an account
- Reports functionality : not terribly useful to capture all of the data if you can’t get some reports out
- Roles/Groups/Users stuff : I have a working version of this but there are a few things that I want to polish before I start using this heavily in the site
- Account/User preferences infrastructure
Not a lot of progress on the 11-05 challenge. I could probably make the case that I’m giving Mike a chance to catch up but that would indicate that I’m a nice guy. I’m not. I’ve been on various vacations and also took some time out to clean out my garage…so we’ll call it a break. I’m now done with my vacations and ready to dig in and get this thing finished. I don’t have a list of action items for this week. I’ll just get settled in and see where things are and return to my regularly scheduled “here’s what I’m gonna do” type of post for next week.
I was able to get the application administration screens knocked out this week. I am almost ready to get an alpha build put on a public server and start handing out some accounts for people to start giving me feedback, there is a LOT more to do but I want to try and get this into people’s hands as quickly as possible.
Tasks for this week :
- new account setup wizard ( carry over from last week )
- ajax support routines need a little love
- finalize deployment strategy and get application onto a public server