Build/design log websites

HomeBuiltAirplanes.com

Help Support HomeBuiltAirplanes.com:

etterre

Well-Known Member
Joined
Aug 30, 2006
Messages
313
Location
St. Louis, MO, USA
Hello to all.
I find myself in a bit of an odd postion: I need to find some sort of "real" project for a web app. I've been noodling about it for a while, and I think I'd like to create an app for a design and build log. I know that there's stuff out there and I've already taken a peek at expercraft.com, mykitlog.com, and myairplane.com I'm kinda curious about what people love (or hate) about these sort of sites. I also know that there's a lot of folks out there that create the pages themselves instead of using an existing templated site and I'd like to hear from them as well.

Here's some of the stuff I have in mind (nerd-stuff last):
  • Something that allows for design... most of the sites I've seen are oriented towards logging build progress and I'd like to log design progress as well.
  • A "new entry" form that allows you to type in some text, add photos and docs, and then add keywords. After submitting the form, a new log entry is added that looks pretty and shows up in the right categories according to the keywords.... My main goal with this form is that the average lazy user (yep, that's me) can go here and make a publicly viewable entry in less than 15 minutes so that I'll actually have a chance of keeping the thing up to date on a weekly basis.
  • Some level of security... I'd like to be able to upload a set of calculations so that I can share them with the engineer "looking over my shoulder" without sharing them with the entire world.
  • Automatic "landing page" so that I can have some sort of comfort level that every reader has read the disclaimer page at some point.
  • Some flexibility in view-types for the reader - ability to read entries from day one in chronological order, "skip the mistakes": read only the entries about parts that made it into the flying airplane, show a tree-view based on the parts of the airplane (Airplane has fuselage, tail, wings, engine, and avionics as children. Each of those has children as well: tail has horizontal and vertical as children, vertical has rudder and rudder trim as children, etc.) so you can choose to read about the entries related to the rudder
  • Database back-end that stores the text and the layout
  • Some pages with SilverLight, just to be cool
  • Any forms will have an AJAX-enabled version
  • Any page that can't be sent to the reader's browser as "pure" HTML will have some sort of HTML "sorry, scripts not enabled page," or (as in the case of the AJAX stuff) a page that offers the core functionality without the shiny stuff.
Thoughts?
 

addaon

Well-Known Member
Joined
Feb 24, 2008
Messages
1,686
Location
San Jose, CA
Silverlight (or flash, or anything more than universally compatible javascript) is a killer for me. "Sorry, you need to install stuff to read about a personal project" means, for me, "I'm more interested in shiny things than sharing information." There's plenty of other information out there without having to wade through software requirements to read a page.

I think some of your requirements are going to be contradictory... 15 minutes to make an entry is fine if it's slapping down some images with a few notes, but you're going to have a hard time making a hierarchy editor (especially with concepts like "still in the design" and "private") that someone trying to do a 15 minute entry is going to use properly. And the only thing worse than no structure is misleading structure. Also, think of the web pages you use... how many of them (that you return to!) only took a few minutes to put together? I guess the question is if the customer is the designer/builder (who wants things easy) or the third party (who doesn't really care what the designer/builder has to say if the presentation is lousy). If the former, why bother with the web?

(On rereading, man, I sound like a jerk above. I think it's a good idea! I've just seen similar projects and their results, and I think that the big problem here is not a technical one, as you seem to have things in hand, but one of clearly stating goals.)
 

addaon

Well-Known Member
Joined
Feb 24, 2008
Messages
1,686
Location
San Jose, CA
Also, the one feature that I don't have at my site that I'd like to have is a place for people to leave publicly-viewable comments. I specifically don't want a wiki or even a threaded discussion form; just something simple. I'll get to it eventually, but for now it's the one thing I miss.

Also also, if you're doing this for design as well as build, you're going to need to solve the Great Unsolved Problem. No one (including google, to my dismay) has put together any decent system for editing and displaying math equations. MathML isn't even universally supported. What I and everyone else I know does is to just hand-typeset things with LaTeX and export to pdf or png (via eps). This is a killer for "15 minute updates", though, and it's also a killer for an all-in-one approach.
 

rtfm

Well-Known Member
Joined
Jan 3, 2008
Messages
3,194
Location
Brisbane, Australia
Roy,
First, well done. As I read your list, I mentally ticked off my personal wish-list, one by one. If you could produce a site like this (I especially like the forms-interface for adding pages) I would be knocking at your door to use it for my own project.

As far as the Silverlight/AJAX enabled pages are concerned, if you can implement an alternative vanilla page - then everyone will be happy. I would.

I think you are on the right track with this. Strength to your arm.

Duncan
 

Mike Armstrong

Well-Known Member
Joined
Aug 17, 2006
Messages
206
Location
near San Diego
Hi Roy.

Sounds great! You have some very good sounding idea's. I will soon be looking for some type of 'build log' and none of the popular ones seem to have overall rave reviews. Whats needed here is feedback from folks that have used some of the currently available build/kit logs to see what is good or bad about them and what is needed if a new log is offered.
 

etterre

Well-Known Member
Joined
Aug 30, 2006
Messages
313
Location
St. Louis, MO, USA
Thanks, I appreciate the feedback!
I'm also on the same page with y'all regarding alternate non-Silverlight/AJAX pages... but unfortunately this is also a "career-enhancing" project, so I have to do some silly stuff so that it looks really shiny so that I look really hip and smart.:rolleyes:

Publicly viewable comments, eh? I hadn't thought about that, but it's a good idea... There's some overhead there, though, with all commenters needing to register, and keeping out the spambots, and giving the owner the ability to delete posts (or approve posts as they come in:speechles)...

The equation editor is also a good idea, but the ones I've used (long, long ago) were always more cumbersome than just writing out equations long-hand. One good example is adding the bounds to a capital sigma for a summation. I'll have to fiddle with that one...
 

addaon

Well-Known Member
Joined
Feb 24, 2008
Messages
1,686
Location
San Jose, CA
Yeah, for comments, it's probably worth integrating someone else's open source package that supplies all the details, rather than doing your own. As someone in a hiring position in the tech world, I can tell you that showing off integration skills is a lot more valuable for most jobs than just showing off development skills. (Also, knowing when to avoid the shiny stuff can get you some jobs at some interesting places. :))

The equation editor is a much bigger chunk of work, because there are unsolved problems there, at least slightly. I don't think either the input method (TeX, other notation) or output method (image, MathML, html-when-possible) is a solved problem. On the other hand, solving it (and open-sourcing it) would not only get you a job, it would make you lots of friends!
 
Joined
Jan 25, 2003
Messages
430
Location
Shirley airport MA
I don’t like the pre made formats offered for builders logs. I built my website by hand myself, its similar to what many other builders do, nothing flashy just easy to navigate through the content. If I want something better in the future I will have it done by a professional web builder. For interaction with other people a lot of builders use Yahoo groups where they set up their own, I’m a member of 120 of them. Either way you will hardly impress anyone with a flashy website until you have real impressive content to put on.
 

etterre

Well-Known Member
Joined
Aug 30, 2006
Messages
313
Location
St. Louis, MO, USA
Yeah, for comments, it's probably worth integrating someone else's open source package that supplies all the details, rather than doing your own.
True, reinventing the wheel is always a bad idea... but in my mind, at least, this entire concept is about wheel reinvention. A builder's log is really just a specialized blog that has extra structure (is this post about assembling the tail, or your insights about the wing ribs, or both?) that allows the reader to find the post that they want. Most of them have a navigation bar that allows you to look at the most recent post or drill down into posts about particular assemblies. Now, from a software-reuse standpoint, it would make a lot of sense to take almost any one of the available open-source blog packages and extend it to add the features that I want. The problem is that I'm a .Net guy and all the ones of which I'm aware are written in java or php. So I'm probably looking to re-invent the wheel just because of language choices... but I haven't yet done the necessary research yet. The more I think about it, the more sense it makes to take a deeper look at either a conversion from, for example, a php based package to .Net (or use the php.net libraries) -or- putting a sort of ASP.Net layer around an existing package...

Hmmm.... It is readily apparent that I haven't gotten beyond paper napkins yet, right? :gig:

Back to basics: goals/purpose/features first (which is still where I'm at), then architecture, build, test, release first iteration, fix bugs, test, release bug fixes, repeat...

As someone in a hiring position in the tech world, I can tell you that showing off integration skills is a lot more valuable for most jobs than just showing off development skills. (Also, knowing when to avoid the shiny stuff can get you some jobs at some interesting places. :))
Agreed on both points - but those sorts of skills don't show up on the resume very well... and most folks seem too focused on ticking off checkboxes to see if you fit the "technical profile" that they're seeking.

The equation editor is a much bigger chunk of work, because there are unsolved problems there, at least slightly. I don't think either the input method (TeX, other notation) or output method (image, MathML, html-when-possible) is a solved problem. On the other hand, solving it (and open-sourcing it) would not only get you a job, it would make you lots of friends!
After thinking about it a bit more, I see your point about the formula editor. I don't think that the answer on the input side is to use TeX (or anything like it) because I just can't see anybody using it for a post to this site - and you guys are my target audience. I'm probably looking at something that gives you and editing box and various child controls for adding summation/integral/root/etc symbols to the text in a more WYSIWYG sort of manner. It would make sense to use a TeX engine behind the scenes, but forcing the user to learn a notation language is a non-starter. At the moment, I'm thinking of a single control that allows you to create a TeX string in a more user-friendly manner, see (and hand-edit) the resulting TeX string, and finally a box to see the output.... On the output side, a combination of html and images is probably the way that I'd go to avoid the "this feature only works on <insert browser name here> version xx or higher" sort of problem. I'd also like to avoid forcing anybody to download stuff (activex, silverlight, flash, whatever) to read the equations - it should be seamless.

Hmm, I've got lots more thinking to do... but I should really get back to remodeling the kitchen. I'll post more here when I have something to show.
 

addaon

Well-Known Member
Joined
Feb 24, 2008
Messages
1,686
Location
San Jose, CA
Agreed on both points - but those sorts of skills don't show up on the resume very well... and most folks seem too focused on ticking off checkboxes to see if you fit the "technical profile" that they're seeking.
Best of luck... those are the types of jobs I always try to avoid. :lick:
 
Group Builder
Top