As to that, I know the square root of fuck-all about subsistence farming (please, nobody put me in a time machine).
4 Likes
Did they include a lab coat and slide rule?
2 Likes
My 85 year old father lives with me… he owned a software company in the 80’s he was more what we would call product manager/owner and not doing development but still more computer savvy than most. You might think this would be an advantage in helping him out with technology, it’s not.
Now I am a software developer but watching him makes me afraid of my future. I try my best to be open to new tech/ux whatever but I find myself getting pissed off when Google/Apple/MS/FB feel the need to rearrange the UI ever 6 months.
8 Likes
As indexing has become more automated, I actually suspect that it has become less important to be familiar with controlled indexing. Eventually I pretty much settled for trying to explain to people that four letter acronyms that are ALSO words in English are bad search terms. And most three letter acronyms stand for multiple things.
2 Likes
People think that librarian is a dying job, but just wait until you need to find some bit of information and have no clue how to begin your search.
“Librarians are the secret masters of the world. They control information. Don’t ever piss one off. ” ~~Spider Robinson
21 Likes
tl;dr: Geezer says “Don’t blame the kids, 'computers” are primitive crap that we will eventually outgrow."
Just my two bits, I’m not claiming any special wisdom, although I’ve been “tricking warm rocks into thinking [1]” for almost forty years now. My first computers didn’t have file systems and we looked things up with proprietary queries that used keys you don’t see on machines any more (Cross of Lorraine anyone?) Essentially everything was in what we would now call a Set rather than a Tree. Kind of similar to the the metadata and search idea in BeOS or the “soup” idea that came much later in in Apple’s NewtonOs. As others have mentioned here Apple has long had this idea of using a Set instead of a Tree structure. The idea being that if things are unique, they’ll be just as easy to find in the set, if they aren’t unique there will only be one of them no matter how many times you refer to it. It just takes fast search that can search by content.
One hard problem that is only starting to improve now, is image and binary search. Those are the types of files that pretty much required a tree until now to make any sort sense unless you’re going to make heavy use of metadata tags. Now with some help from trained pattern recognizers, there’s some hope of a local search like “photo of me holding a Timex Sinclair ZX81 next to one of the Front End Processors in the TPF.”
Knowing how to make a PC work is probably about as useful to a kid who will reach adulthood in ten to fifteen years as me knowing how to IPL an AS400. PCs (I’m including Mac and Linux when used as personal computers here) will still be around, but day to day life will likely mean interacting with a single interface across all the compute infrastructure that is available to you, probably though whatever replaces smart phones, maybe glasses, maybe voice and whatever display is nearest. Who knows?
Sure CS folks will need to understand Stacks and Linked Lists and b-trees and practice evocation and banishments and read their daily Knuth [2], but those will matter to the average user about as much as they do now. Knowing more will likely allow one to understand how to do things better, safer and more efficiently for a good long time to come, just like knowing a little bit about electricity will keep you from doing an unpermitted rewire of your house with 30 AWG, and, for fun or profit, they may solder up some cool prototype or toy with that knowledge too. Most folks just want the light switch to work.
5 Likes
The french national library has a great digital library. But the digital material doesn’t have call numbers, they’ve been stripped out-- it’s full text search only, and if your subject doesn’t have a unique word associated with it, well you’re stuck with the problem of figuring out a cloud of terms that authors might use when discussing a particular subject.
Imagine trying to research techniques for painting with oils (Dewey 751.45436), and in order to avoid a lot of results about petroleum extraction (Dewey 622.3382), you were forced to add “palette knife” and “easel” to your searches.
5 Likes
<pedantic>
Except they aren’t.
</pedantic>
2 Likes
I have an antique slide rule. One of these days I’ll learn how to use it.
5 Likes
The 001.9 bit bucket keeps growing/sub-dividing.
3 Likes
May I recommend ISO 8601 for dates, and at the start of the file name not the end. Then no matter the modification or creation date, you can file name sort in the directory according the the date you stipulate. Saves me loads of time.
You’ll like this. I’ve not read it yet but the reviews tell me I will.
6 Likes
in every user interface study we’ve ever done […], [we found] it’s pretty easy to learn how to use these things ‘til you hit the file system and then the learning curve goes vertical.
I really cannot relate to this. What’s so difficult about a directory tree system?
As an old school (first used a computer - an IBM mainframe VM system - in 1985) user who not only knows how to organise my files, but INSISTS on it, this is why I refuse, point blank, to use on my Mac any Apple software that tells me I don’t need to worry about where my files are, the app will take care of that. NO, they are MY files and I fucking well want to know exactly where they are. It may never have occurred to these fuckers that I might want to use those files with a DIFFERENT app, so up their own arses were they.
Where I have, after all, decided to use such an app (e.g. iTunes - yeah, so mock me - I use it and have wrangled it into submission to manage my music files the way I want them managed, and stopped at v12.3 - the last usable version before Apple started removing features I liked) I only did so by forcing it to use MY file locations and organisation, not theirs.
You don’t keep your music in the file system, that would be crazy. You keep it in this app that knows about music and knows how to find things in lots of different ways.
I am not crazy and I do keep my music in a file system, and the point is that it is me who knows how to find it not an app! I don’t need lots of different ways to find any music, thanks very much.
I do find it odd that kids allegedly do not get tree structures. It’s just sets and subsets and I thought schoolroom maths these days included sets, whereas in my day it was mostly all just arithmetic and geometry. I mean, how do they not get:
Music vs Videos
Music: Artist A vs Artist B
Music: Artist A: Album 1 vs Album 2
I can’t think that I’d ever want to search like that, but I do often want to find a photo of a certain ‘type’ (e.g. very generic subject) that I took e.g. ‘around the middle of last year’. Having all my photo folders in ‘types’ and all folders dated within that, and all files dated - and all dated in YYYY-MM-DD-number-maybeaname style, means I can home in on a photo pretty damn quickly without EVER having to tag a single one, which I have no intention of starting to do. I really dislike the idea of tags - how can I ever know in advance - at the point I create or file the photo - what element of the picture may be one I wish to search for in future? And it is so much extra work!
All very fine if it works for others, but I guess I’m really so old school that I intuitively have a mental map of my file organisation, its methods, and why and how I deliberately made it like that - and that’s more than good enough for me and means I really think I don’t need tags.
(I appreciate you were taking about intelligent analysis of image content rather than tags, but tags is today’s ‘image content’ reference, and if I can’t reliably get tags right, I am skeptical of an AI identifying the things I might want to look for.)
And as for
Same with photos: we’ve got an app that knows all about photos. And these apps manage their own file storage. […]
Their app knows fuck all about many aspects of my photos and I’ve got better things to do than educate it. Apple’s Photos is a particular offender and has been forbidden to open any photos on my Mac unless I specifically open the individual file with that app.
(Probably needs an /rant tag.) 
7 Likes
I suspect that it is by necessity, rather than desire; but in my (thankfully relatively infrequent) contact with electronic discovery tools there is a lot of emphasis on search capabilities there as well.
It’s obviously preferred if the data storage structure is nice and sane and someone can just pull exactly what is being asked for; but when that fails(given the purpose of electronic discovery, not infrequently because the person who would have known is now no longer with the company and on the other side of something unpleasant that Legal prefers not to discuss in detail) you can pull out the “every email with X in its subject line between date a and date b”, or “every email sent to person X, and only person X, between a and b”; or “all documents classified ‘sensitive’ authored or edited by person X”; or any other variation.
In some cases handling things entirely via search parameters is even preferred, because it makes it easier to make statements about completeness (“here is the package of results, here is the query that produced it, here are the retention policies applied to the locations the search covered” vs. “the person who knows that found this and thinks that covers it”) and to keep the information at arm’s length: normally only a very high privilege account, likely controlled by IT, would be capable of scouring all mailboxes in the company, or every file storage location; but neither HR nor Legal really wants someone from IT poking through material that might be ruled to be outside the requirements of discovery, or sufficiently unpleasant as to be exactly why discovery is happening; so just constructing the query and then blindly handing them the results is preferred to doing any looking by direct inspection.
I just started a selective re-reading of Jef Raskin’s The Humane Interface - in which he discusses the human-machine-interface thinking that went into the early conception of the Macintosh system (that he helped create), and some ideas that have never made it into production, but are still fascinating and (imo) worth re-visiting by current UI designers. One of the coolest ideas that he developed is the Zoomworld concept, a modeless GUI where users fly around the system like google earth, zooming in and out to greater levels of details.
I also blame Microsoft for creating weird inner loops in the hierarchy that obfuscate the structure, (esp. for those who aren’t already savvy to the model). Those symlinks and search functions are certainly helpful, but not having a sense of the underlying structure feels dis-empowering, not just in terms of an abstract knowledge of the tech - but also in the ability to wield those tools in an independent manner. The real shift that is driving this though - is the generalization of stored assets to a level that is not one-to-one with the storage mediums and specific systems, whether cloud or distributed NAS etc. For storage, I think that is an inevitable direction that things will grow toward - just because having 13 different hard drives in 6 different computers and various removable media becomes more and more unwieldy.
One of the things that was revolutionary about the graphical windowing interface environment was the idea of discoverability - that a user could just “poke around” and figure what the system could do without a manual. Something that is less possible in a command-line paradigm. Searching just doesn’t offer that same ambient presentation of context that informs about a systems organizational form. Maybe that’s fine for systems where the contents organizational context isn’t important - but in many cases it is (like books in a dewey decimal library).
I recognize the value of both prescriptive and descriptive perspectives, and the whole kids-vs.-olds aspect of this - but we are not Uatu The Watcher - and we’re all a part of the ongoing techno-cultural developmental conversation to some degree. When we lose contextual visibility, the likelihood of serendipitous association drops - that always feels like a lost opportunity to me.
3 Likes
It IS disempowering. I bet that most of these ‘kids’ if educated about file systems would find at some point a large penny dropping and suddenly they’d ‘get’ so much more about how their system works and where their data is and how much more usable to them it just became.
And the ‘where their data is’ issue may turn out to impact another issue. Kids seem often to not care about data privacy and perhaps it is in part because they have no idea where their data is. Having greater abilities to control it, by knowing where it is, might make a difference - an empowering difference (or I may be a naive optimist).
9 Likes
Very true. Hence the office expert on Excel, or Word who has no idea how to use another program, or what to do if something unusual appears on screen.
1 Like
I think this is some that is really tough to imagine when one has already developed the understanding - but there are a number of folks who I’ve met - people who have been using computers as long as I have (since the late 80’s early 90’s) - (though much less often and not professional settings), who still struggle with these concepts. My best guess is that it is something kind of analogous to the developmental leap that a toddler makes re. object permanence - the realization that there is a “hidden” level of continuous reality that is momentarily being variously represented. It’s that sense of an underlying integrity that forms a common factor between the GUI view of a folder and the command line path, and something that can be “transported” on portable media - etc. that if not directly experienced at some point - it becomes tough for some folks to form a facile, flexible relationship to the mediums - and yet they can still operate with them under tightly defined contexts.
3 Likes
Dewey’s hierarchy contains plenty of assumptions.
Sure, it’s useful to be able to browse a bookcase and find lots of related information, but does that bookcase really need to be amid the bookcases on social deviance?
2 Likes
oof - yeah, ok not the best example.
