SueVT said:In programming, apostrophes and quotation marks are often used internally to indicate an absolute or interpreted value. So the search would need to convert the input text to an internally-neutral format to correctly form the query. The "offending" characters, like apostrophes, are wrapped in more apostrophes, or preceded by "escape" characters like \.
Thanks Sue! Exactly!!
In case ADS technical team members or other database aficionados are watching this thread, some databases solve related problems by creating quiet (hidden) search strings with multiple permutations, such that a search for Angel's would automatically search for "Angel's" OR "Angels" OR "angel's" OR "angels" (avoiding issues of punctuation and capitalization), and the most useful databases will give advanced users an option to specify exactly what permutations they want and in what combinations (using the standard AND, OR, NOT, and other Boolean operators). The advanced search feature in the daylily database is reasonably good for a WYSIWYG option; but since I do a lot of database searching in my professional life, I'd deeply appreciate something more advanced. I'd love to be able to choose whole groups of color terms: "purple" OR "lavender" OR "lavendar" (in case of misspellings) OR "orchid" OR "violet" (etc.). I'd also love to be able to use truncation (wildcard) searching: poly* would return polychrome, polymerous, and anything else beginning with poly. I actually often will do a google search in addition to my search of the database because of the differences in the underlying algorithms.
I think these could be important and useful features. For example, let's say I was considering to register a late purple, and I wanted to see what was already out there (ML to VL). I would have to run dozens if not hundreds of searches through the existing advanced search option (on ADS or via the garden.org mirror site) to try out each of the potential locations (color in description, description keywords) for a color word and all the permutations I could think of that a registrar might have used. The more complete I wanted to be (i.e. the more carefully I wanted to curate my registration), or the more complex my seedling's attributes, the more work I would have to do and the longer it would take. If I now have to do this x 100 seedlings, well, you can see why it might be nice to build thorough chains of search terms and run a single search with ALL the various options. The really nice thing about this approach is that hybridizers could build their key search for each of their programs or goals (and for their surprise sdlg keepers), keep these searches saved in Word or another text document, and then just copy and paste each search into the search engine every year after the new registrations are processed to see what else has grown up in the meantime.
I also want to emphasize that, given the options of NOT having registration or databases for daylily hybrids, the ability to access and search an online database on more than just a few features is a remarkable accomplishment, and something I value quite highly. Perhaps bulk discounts for hybridizers who register a lot (to encourage registration), and more automated systems for updating the database (hybridizer password-protected log in with moderated editing privileges, certified superusers who can modify records for the older archive) could address the challenges raised by others.
Given how expensive both programming and data management can be (I've had to budget for these before), the registration fee is quite reasonable. We should acknowledge that changes to any registration process or database upgrade would have to be supported somehow, whether this is through volunteer activities, registration fee increases, grant seeking, targeted fundraising, or other activities. I assume ADS is a 501(c)(3), which could open up other possibilities.
The other important consideration for any potential change to registration or database or both would be a formalized process—for example, to collect ideas, create a "straw man" proposal of changes, allow for open comment and a transparent process to modify this proposal, have some process by which the proposal is finalized, identify the method for change (resource support, timeline, etc.), and allow for a substantial post-hoc debugging period. I mention this only because change is a process, it's hard, and I want to show respect for the effort it would take as we all throw out our wish list features here. So: thanks ADS database team!
And thanks to everyone for this fun thread!
Meghan
