Error messages like this (from our corporate travel booking system) make me think, “But why? Why can’t I pick the options that actually describe what I need?” (What I really wanted is in the drop-downs.)
Didn’t someone designing this stop and think, “What will users needs to do while interacting with this application?” It’s a small thing but reveals such a user-oblivious mindset.
{ 7 comments }
Well, as a programmer, this problem makes a lot of sense. Building in the ability to mix “Anytime” with a specific time is expensive. It is likely that the client, not the programmer, made the final determination that customers didn’t need this “feature,” to keep costs down.
Midway through a project, clients often ask for a feature-by-feature estimate of the cost of completing the project so they can find savings. I have often been astonished at the decisions clients make to save money. They often end up costing them quite a bit more, as adding a feature after the program has been released is much more expensive than building it into the software during it’s initial development.
I would estimate that adding a feature to allow for mixing “Anytime” with a specific date will cost two to three times more today, than it would have cost during the initial development cycle.
Why would it have not been included in the first place, based on a simple observation of user behavior – people don’t have the same time requirements for all flights on an itinerary?
You can ask it to show you outbound flights with 2 hours of 3 PM and inbound flights within 2 hours of 7 AM, outbound flights within 2 hours of 6 PM and inbound flights within 2 hours of noon, but you can’t ask it show you outbound flights for the whole day and inbound flights within 2 hours of 5 PM? That’s just dumb.
Plus: when you get to the results, you see the flights in the selected interval… along with a slider that lets you change the time interval on the fly, with instant expansion/contraction of your results list (looks like Ajax). So all the data is in fact there, the selections I showed are just filtering a data list and on the results page can be instantly adjusted.
I generally use the Continental.com site when booking personal travel and it handles this same issue much better by giving you wider time internals like “morning” and “midday” and displaying results more predictably. Given how long people have been designing web interfaces for querying flight data, this seems pretty silly to me.
I totally agree with you, John. But I can see some bean counter at corporate learning that including this functionality would cost $1200 and saying — hey, our customers can do without that!
And in that vein I guess my larger “WTF?” moment is why a private company (and John correct me if I’m wrong, but not a very large company at that) reproduces a travel website. Why not just have users go to Expedia or Orbitz or whatever, and if necessary have someone double-check that yes, the flights the user chose are appropriately priced.
Oh, it’s not ours as in we built it – it’s the corporate version of Expedia, which we use. So, yes, this is the product of one of the biggest names in the market.
(The corporate version is basically very similar, but incorporates approvals, central billing, etc.)
Ah. Well then that is ridiculous, firstly because a corporate version of Expedia is going to be used by a lot of people, so the added coding cost will be spread among a number of users — but even more so because the “lay” version of Expedia does allow you to do the “any” out / specific return search that you described. (I know, I just checked.)
Comments on this entry are closed.