Windows explorer and creating a filename which gets sorted /after/ the letters ?

"mick" wrote

| Interesting. I guess it never occurred to me to name
| a file or folder starting with a number. I can't imagine why I would
| want to.
|
| I do it all the time both for files and folders when sorting photos.
| e.g. yearly folder named 2019, then sub folders 01 January, 02
| February. 03 March, etc.
|

But that's backward, unless you live in Europe. If
I had such a folder I'd probably call it Jan-1-19. I also
don't sort photos by date but by topic. Naming a photo
010120191 is not any better than the camera naming it
P10533492. So I'd never have a folder named Jan-1-19.
I have folders on my Graphics drive, in the Photos
folder, with names like Personal, Work, NASA pics, Roses,
etc. The image files in those folders, if I decided they were
things I wanted to keep, were also renamed with meaningful
names, like "NYC 3". If I name that for a date that I went
to NYC I'll have no idea what it means.

Occasionally I might have something with a number name,
but in general I don't. So in general, if I have something
I want at the top of the folder, I can just name it AA*.

I actually do name my website server logs. Tue10-1.
Wed10-2. Etc. Those then go into a folder named
stats Oct 2019. I guess I do that because the date
is always a secondary qualifier. The main point is that
the folder contains web server logs. Stats.

On Sat, 5 Oct 2019 02:04:18 +0100, "J. P. Gilliver (John)"
wrote:

In message , Mayayana
writes:

Agent Ransack puts a context menu in Explorer, is extremely
fast, and doesn't index. As far as I know, that's the main
difference between the two: Everything indexes and has a
confusing name. AR doesn't.

I think they're very different beasts; Everything only searches on
filenames - it doesn't do content at all. (I'm not sure whether AR does
filenames.)

Of course AR can search on filenames. Why wouldn't it?

On 2019-10-04 8:40 p.m., Char Jackson wrote:
On Sat, 5 Oct 2019 02:04:18 +0100, "J. P. Gilliver (John)"
wrote:

In message , Mayayana
writes:

Agent Ransack puts a context menu in Explorer, is extremely
fast, and doesn't index. As far as I know, that's the main
difference between the two: Everything indexes and has a
confusing name. AR doesn't.

I think they're very different beasts; Everything only searches on
filenames - it doesn't do content at all. (I'm not sure whether AR does
filenames.)

Of course AR can search on filenames. Why wouldn't it?


They are both excellent at what they do, I use them both according to
needs. Everything for filenames, Agent Ransack for content.

Rene


Rene

J. P. Gilliver (John) wrote:
In message , Mayayana
writes:
"Frank Slootweg" wrote

| Have you actually _tried_ the "Everything" utility?
|
| No, I haven't. I didn't know I was supposed to try it! :-)
|

Some things seem to congeal their own religion. "Everything"
is one of them. Malwarebytes is another. I like Agent Ransack.
Then, of course, there's the ultimate have-to: "What do you
mean you can't read docx?"

Agent Ransack puts a context menu in Explorer, is extremely
fast, and doesn't index. As far as I know, that's the main
difference between the two: Everything indexes and has a
confusing name. AR doesn't.


I think they're very different beasts; Everything only searches on
filenames - it doesn't do content at all. (I'm not sure whether AR does
filenames.)

I agree Everything's name is _awkward_, as it makes it difficult to talk
about it; I don't think it's any less _confusing_ than Agent Ransack.

Agent Ransack walks the file tree, on demand.

It might take a minute or two for a filename search, depending
on the size of drives, and whether they're SSD or HDD.

Agent Ransack also has content search capability.
You can do a wildcard search, like you might in WinXP Search

filename = *.bat
content = "echo" "find all the BAT files with an echo line"

It can open a content search thread per CPU core. If you
had a 4C 8T processor, it would open eight threads,
and it could be searching inside eight 1GB text files
at the same time. This speeds up the content search,
compared to say, WinXP doing a Content Search without
any indexing being involved. When a core becomes available,
the next file needing a content search is queued up.

In the above example, it would only be queuing up .bat
files for the multi-threaded content search phase. If you
restrict the area-to-be-searched to C:\scripts, then
only a tiny section of the disk drive need be searched.

To do a content search of an entire hard drive with
Agent Ransack, takes as long as it takes to read the
entire drive. It could be I/O bound, if you have enough
cores. A large drive is going to take a couple hours.

Agent Ransack supports regex (Regular Expression)
entries for search. A search for "^fi" would return
only filenames that begin with the letters "fi". If you
leave regex switched on, you have to use escape characters
to prevent punctuation from being interpreted as regex
commands. Perhaps "\^fi" would search for any filename
containing the literal three letters "^fi". So if a
filename was "carrot^fi.txt", that would be a match.

*******

Everytning.exe is a filename search only. It is designed
to do one thing well, and that's filenames.

It reads the $MFT directly, generating a file list.
I think it was the novelty of doing that, which
spurred the developer on. It looks like salting the
filelist with timestamps and file sizes, reduces the
efficiency of this step (it takes a lot longer to
add that info, than the filename step).

myfile.txt Tuesday Oct.12 123,456 bytes

Once the initial scan is done, then at least for NTFS,
the USN Journal is used to keep track of file creations
and file deletions, and then everything.exe would amend
the file list (in real time) with the information. This
allows (within a few seconds) for newly created files
to show up in a search.

Obviously, the same things cannot be said for FAT32. FAT32
is not as feature-rich as NTFS, and the design of the
file system does not allow Everything.exe to have fresh
info about your FAT32 newly created files. You'd have to
re-scan the FAT32 volume, to refresh it.

Agent Ransack, by comparison, is file system agnostic.
Agent Ransack requires no knowledge of the
file system type. As long as a file system can provide
a findnextfile capability (tree walk), and allows reads,
Agent Ransack should be happy.

Everything.exe can find your filename in one second.
Agent Ransack can find your filename in two minutes.
Windows Search (indexing enabled) can find your filename in one second.
Windows Search allows programmatic search (I can call it
from a script!), but hardly anyone does that. I found
some code to do that, on the web. No, you won't like it.

Paul

"Paul" wrote

| Everything.exe can find your filename in one second.
| Agent Ransack can find your filename in two minutes.

You're being theatrical.

My XP C drive is using about 6 GB. I just did a search for
scrrun.dll. It fould that plus scrrun.dll.mui before I could
start counting. I then did a search for a more obscure file.
A config file for my own software that's buried a few layers
down. Since it's my own software I also have a number of
copies in my coding folders. AR found all 9 copies in what
I'd guess was slightly over 1 second. If I give it a file name
that doesn't exist it takes maybe 2 seconds to do a full
C drive search.

That's why I started using AR. It's extremely fast. It will
also look in all kinds of files. I occasionally want to search
for text in a CAB because CABs store a plain text list of
their content, so it's easy to find abcde.drv in the Windows
system file CABs. Windows search can't do that. (One thing
AR lacks that would be nice is a list of exemption folders. For
instance, on Vista+ there's no sense having it search winsxs.)

Of course it can't be so efficient with newer, ultra-bloated
systems. Nor can AR make up for bad housekeeping.
Someone who doesn't create data partitions and who
has a 2 TB C drive packed with photos they don't want,
downloads they don't need, email they never cull, and
software they don't use but don't remove, will not be
able to find things as well.

For most things I don't need AR because I know where
I put things. If I do need AR it's usually folder search.
Software installers are in Software on my "Back40" K drive.
Security articles are in Security on my "Attic" J drive.
Programming docs are either in the Code folder in the
Attic, or in Programming Info in the "Annex" D drive. My
customer list is in Work Files, in the Attic. If all else fails
I check the "Closet" I drive, where I sometimes
dump duplicates.

I also use a shell extension that mimics the old Win98
feature that allowed for custom links on the left side
in folder windows. For most things I do, a single click
brings me to the folder I want. I like to have it set up
like my cellar workshop: Everything has a place and every
cabinet has a label.

But I can see how Everything might be good for those
people who never delete a photo and name them with dates.

I used to frequent the photo newsgroup and was struck
by how many people use file organizer software. That was
a frequent topic. Most people are so disorganized and have
so little familiarity with the file system that software is
now designed to find their files. I think it started with Picasa,
presenting a non-locational list of image files. A entire extra
level of abstraction for the file system. People like
that don't even know where their files are, much less organize
them. And that's probably the majority of people. They're the
same people who talk about "work flow". They don't use their
computer so much as they use software that presents them
with a specific conceptual model of their computer.

But actually, that started many years ago with MS Word,
storing everything in the personal docs folder without asking.
When you ask these people where their tax records are they
say, "I don't know but Word knows". Their photos of last
Summer at the lake? Who knows?! "Photoshop knows where
they are. That's all that matters.". Backup? Who knows.
"Aconite handles that." Or, increasingly, it's
all online anyway, at someplace like Google Docs or Adobe
CS web storage. So their computer is really just a kiosk
interface to services. It gets further and further away from
actually creating and managing files, with people paying for
layer upon layer of training wheels to allow them to actually
use their computer productively.

In article , Mayayana
wrote:

I used to frequent the photo newsgroup and was struck
by how many people use file organizer software. That was
a frequent topic. Most people are so disorganized and have
so little familiarity with the file system that software is
now designed to find their files. I think it started with Picasa,
presenting a non-locational list of image files. A entire extra
level of abstraction for the file system. People like
that don't even know where their files are, much less organize
them. And that's probably the majority of people. They're the
same people who talk about "work flow". They don't use their
computer so much as they use software that presents them
with a specific conceptual model of their computer.

they know where their files are, it's that they don't want to manually
sort hundreds of thousands of them instead of what they really want to
do, which is create great photos, videos, etc.

But actually, that started many years ago with MS Word,
storing everything in the personal docs folder without asking.
When you ask these people where their tax records are they
say, "I don't know but Word knows". Their photos of last
Summer at the lake? Who knows?! "Photoshop knows where
they are. That's all that matters.". Backup? Who knows.
"Aconite handles that." Or, increasingly, it's
all online anyway, at someplace like Google Docs or Adobe
CS web storage. So their computer is really just a kiosk
interface to services. It gets further and further away from
actually creating and managing files, with people paying for
layer upon layer of training wheels to allow them to actually
use their computer productively.

it's not training wheels. it's using very powerful tools to make users
*more* productive because they're no longer doing grunt work.

think of it as an assistant who works incredibly fast and never gets
tired.

In message , Mayayana
writes:
"mick" wrote

| Interesting. I guess it never occurred to me to name
| a file or folder starting with a number. I can't imagine why I would
| want to.
|
| I do it all the time both for files and folders when sorting photos.
| e.g. yearly folder named 2019, then sub folders 01 January, 02
| February. 03 March, etc.
|

But that's backward, unless you live in Europe. If

No, the US way is backward, or at least illogical (-:

I had such a folder I'd probably call it Jan-1-19. I also

I don't think Mick meant he had a folder for 01 January, just that he
had one for January, which he _named_ 01 January to make it (and the
other similarly-named month folders) appear in order. (The names of the
months aren't in alphabetical order.)

don't sort photos by date but by topic. Naming a photo
010120191 is not any better than the camera naming it
P10533492.

Certainly no _worse_, though, and I'd submit it is better in that it
conveys more information than the P format does. (Though in that format,
with the digits all run together, I'd have to know the format before I
could discern the information.)

So I'd never have a folder named Jan-1-19.

I _do_ have folders named something like 2019\10\5, though not in my
images area.

I have folders on my Graphics drive, in the Photos
folder, with names like Personal, Work, NASA pics, Roses,
etc. The image files in those folders, if I decided they were
things I wanted to keep, were also renamed with meaningful
names, like "NYC 3". If I name that for a date that I went
to NYC I'll have no idea what it means.

No, but "2018 NYC" wouldn't _hurt_. But each to his own!
[]
I actually do name my website server logs. Tue10-1.
Wed10-2. Etc. Those then go into a folder named
stats Oct 2019. I guess I do that because the date
is always a secondary qualifier. The main point is that
the folder contains web server logs. Stats.

However, if you named them "10-1Tue", they would by default list in
order, at no extra effort.

--
J. P. Gilliver. UMRA: 1960/1985 MB++G()AL-IS-Ch++(p)Ar@T+H+Sh0!:`)DNAf

"You _are_ Zaphod Beeblebrox? _The_ Zaphod Beeblebrox?"
"No, just _a_ Zaphod Beeblebrox. I come in six-packs." (from the link episode)

In message , Mayayana
writes:
"J. P. Gilliver (John)" wrote

| I think they're very different beasts; Everything only searches on
| filenames - it doesn't do content at all. (I'm not sure whether AR does
| filenames.)

Oh. I didn't know that. Part of the reason I stopped using
Windows search was because it was so bad at searching
for content, which is what I do most. If I know the filename
I don't usually need to look for it. I know where I would have
put it.

Very commendable. I too _try_ to sort by content/category, but sometimes
there are things that get in the way of that - such as librarian's
dilemma, of which more later. (One example is the desire to retain the
original filename, so if I encounter it subsequently I can see if I've
already got it. [Of late, I've been adding something descriptive in
[this] sort of bracket, which isn't often used in original filenames.])
[]
| I agree Everything's name is _awkward_, as it makes it difficult to talk
| about it; I don't think it's any less _confusing_ than Agent Ransack.

Agent Ransack is a unique name, even though it might
be an odd one. Everything means everything. The only worse
name would be, perhaps, "Something" or "The Other Thing":

"Have you tried Something to search for files?"

That's what I meant by being difficult to talk about, and why I wish
they'd chosen something else. Purely on uniqueness, I think they're
about even: I'm not aware of anything else (I was going to say any other
software, but actually _anything_) called "Everything". [IIRR the UK
telecomms company called EE actually say it stands for "Everything
Everywhere", but I suspect even most of their own employees don't know
that; certainly most people just refer to it as EE, and that's what it
says on their storefronts.]

2
--
J. P. Gilliver. UMRA: 1960/1985 MB++G()AL-IS-Ch++(p)Ar@T+H+Sh0!:`)DNAf

"You _are_ Zaphod Beeblebrox? _The_ Zaphod Beeblebrox?"
"No, just _a_ Zaphod Beeblebrox. I come in six-packs." (from the link episode)

In message , Char Jackson
writes:
On Sat, 5 Oct 2019 02:04:18 +0100, "J. P. Gilliver (John)"
wrote:

In message , Mayayana
writes:

Agent Ransack puts a context menu in Explorer, is extremely
fast, and doesn't index. As far as I know, that's the main
difference between the two: Everything indexes and has a
confusing name. AR doesn't.

I think they're very different beasts; Everything only searches on
filenames - it doesn't do content at all. (I'm not sure whether AR does
filenames.)

Of course AR can search on filenames. Why wouldn't it?

I didn't mean to imply it didn't; I just didn't know. (I think
Everything is (sometimes, at least) a lot _faster_, though, at that
task.) [I have both, though I use E more - I rarely search on content,
mainly because it _is_ slow - can't be otherwise of course. When I do
want to, AR is indeed the one to use.]
--
J. P. Gilliver. UMRA: 1960/1985 MB++G()AL-IS-Ch++(p)Ar@T+H+Sh0!:`)DNAf

"You _are_ Zaphod Beeblebrox? _The_ Zaphod Beeblebrox?"
"No, just _a_ Zaphod Beeblebrox. I come in six-packs." (from the link episode)

In message , Mayayana
writes:
"Paul" wrote

| Everything.exe can find your filename in one second.
| Agent Ransack can find your filename in two minutes.

You're being theatrical.

My XP C drive is using about 6 GB. I just did a search for
scrrun.dll. It fould that plus scrrun.dll.mui before I could
start counting. I then did a search for a more obscure file.
A config file for my own software that's buried a few layers
down. Since it's my own software I also have a number of
copies in my coding folders. AR found all 9 copies in what
I'd guess was slightly over 1 second. If I give it a file name
that doesn't exist it takes maybe 2 seconds to do a full
C drive search.

Everything doesn't actually _do_ searches: what it actually does when
you type in what you're looking for is filter. It has already _done_ the
search, and found everything; when you type the target, it filters what
it already knows as you type. (Aha! Maybe that's where the name comes
from after all.) Granted, it _does_ have to do the original
look-for-everything, and that _does_ take finite time, but it doesn't do
it often, and it keeps its list-of-everything up to date. I've just
opened it - for some reason I didn't already have it open, as I usually
do - and it did "Updating database" for about 20-30 seconds. It's
telling me "402,678 objects" in the status line at its bottom. I'll now
try your example: It filtered as I type, and as soon as I'd typed
"scrr", it's showing "14 objects", being ten copies of scrrun.dll, and
four scrrun.dll.mui. As soon as I type a z (so it says scrrz in the
box), that changes to "0 objects".

That's why I started using AR. It's extremely fast. It will
also look in all kinds of files. I occasionally want to search

For looking _inside_ files, I think we're in agreement in these 'groups
that AR is the best of its kind. (Similarly, most of us agree Everything
is best for searching by filename.)
[]
systems. Nor can AR make up for bad housekeeping.
Someone who doesn't create data partitions and who
has a 2 TB C drive packed with photos they don't want,
downloads they don't need, email they never cull, and
software they don't use but don't remove, will not be
able to find things as well.

(-: [My C: partition (this is W7) has 40.1G used, and that's almost
entirely Windows itself, installed software, and the software's own
housekeeping (configuration files etc.); I never _save_ anything to C:.]

For most things I don't need AR because I know where
I put things. If I do need AR it's usually folder search.
Software installers are in Software on my "Back40" K drive.
Security articles are in Security on my "Attic" J drive.
Programming docs are either in the Code folder in the
Attic, or in Programming Info in the "Annex" D drive. My
customer list is in Work Files, in the Attic. If all else fails
I check the "Closet" I drive, where I sometimes
dump duplicates.

I don't have as many _drives_ (this is a laptop for a start, although I
could have partitions), but I do have _folders_ (images, genealog.y,
movies, sounds&mu.sic, for example). (Each with many subfolders of
course.)
[]
But I can see how Everything might be good for those
people who never delete a photo and name them with dates.

I used to frequent the photo newsgroup and was struck
by how many people use file organizer software. That was

What I in my previous post referred to as librarian's dilemma is
particularly prevalent with photos. Let's say you have a folder for the
Smith family, and one for Fido Jones your dog; where do you file a photo
of Fido with the Smiths? You can't make Fido a sub-folder of Smiths if
he isn't part of that family. A lot of photo software gets round this by
making "albums", and any single photo may be in more than one album; it
doesn't (except for the crassest of such software) create multiple
copies, it does it by using "tags", which are in the "albums" which are
lists maintained by the software. I don't like this because I like to
know where my files really are, and don't use such software; however, I
do acknowledge the problem. (Another problem with the album approach is
that the list formats are probably proprietary to each software, and
maybe even to a limited range of versions of that software.) It doesn't
apply just to photos, though - there are all sorts of reasons why a file
might qualify to be put in more than one place. (I suppose you can set
pointers/links in one of the places, but that's tedious - plus, you are
likely to forget them if you ever actually move the real file, or just
tidy the structure of the key it's in.)
[]
level of abstraction for the file system. People like
that don't even know where their files are, much less organize
them. And that's probably the majority of people. They're the
same people who talk about "work flow". They don't use their
computer so much as they use software that presents them
with a specific conceptual model of their computer.

Playing devil's advocate here (I'm more like you), I could argue that
that's still "using their computer". (I could also say that it's a
matter of degree: _you_ [and I] are happy with the "conceptual model"
the OS presents us with of our files and folders, rather than keeping
track of where the individual data clusters are on the "drive".)

But actually, that started many years ago with MS Word,
storing everything in the personal docs folder without asking.

Yes, I found the _obscurity_ of that irritating.

When you ask these people where their tax records are they
say, "I don't know but Word knows". Their photos of last
Summer at the lake? Who knows?! "Photoshop knows where
they are. That's all that matters.". Backup? Who knows.
"Aconite handles that." Or, increasingly, it's

I was going to say (playing DA again) why should it actually _matter_
that they don't know where the files "are", any more than it matters
what the chemical composition of the fuel you buy to put in your car
(other than whether it's Diesel or "gas")? Whether you know that or not
doesn't affect how well it goes or how you drive. [OK, minor faults in
the analogy - don't over-analyse, I can't be bothered to think of a
better one at the moment.] I would have answered myself with "if they
don't know where their files are, they're probably not backing up
properly", but then you mentioned Aconite (which I presume is some sort
of backup software); if that works with Word, Photoshop etcetera and
actually does backup properly, then perhaps they really _don't_ need to
know where their files really "are".

all online anyway, at someplace like Google Docs or Adobe
CS web storage. So their computer is really just a kiosk
interface to services. It gets further and further away from
actually creating and managing files, with people paying for

They'd argue - if they think about it, which they almost certainly don't
- that creating and managing files isn't what they've got a computer
for: it's creating and managing pictures, documents, etcetera. The fact
that these are "files" - or collections thereof (e. g. for web pages) -
is probably of supreme indifference to them.

layer upon layer of training wheels to allow them to actually
use their computer productively.

(I'm finding your analogy beginning to creak: I'm trying to picture what
layered training wheels would look like in practice! But I know what you
_mean_. [And I agree, of course. But I'm old, or getting that way.])

2
--
J. P. Gilliver. UMRA: 1960/1985 MB++G()AL-IS-Ch++(p)Ar@T+H+Sh0!:`)DNAf

"You _are_ Zaphod Beeblebrox? _The_ Zaphod Beeblebrox?"
"No, just _a_ Zaphod Beeblebrox. I come in six-packs." (from the link episode)

"J. P. Gilliver (John)" wrote

| I'm not aware of anything else (I was going to say any other
| software, but actually _anything_) called "Everything".

Everything is called everything. Considering you Brits
invented the language, you sure do seem to have
trouble with it.

"Mayayana" on Fri, 4 Oct 2019 21:37:34 -0400
typed in alt.windows7.general the following:

| I do it all the time both for files and folders when sorting photos.
| e.g. yearly folder named 2019, then sub folders 01 January, 02
| February. 03 March, etc.
But that's backward, unless you live in Europe.

Or lives in Europe long enough to confuse things. Unless
critical, I still give my birth date as "five five" and it maps in
either convention. Otherwise, I say "Mai Fifth" and let them enter
the proper numbers in the proper place on the form.
Date conventions are one of those things which have lots of
"installed user base" and what works in the written world doesn't work
as well in computers. Humans can sort out that "the 3rd of June"
comes out after "January 2nd" and both are before 1 July.
_A_ problem with naming files "Jan1" is that then next two files
are Jan11 and Jan15, not Jan2 and Jan3, And of course, the months
sort April, August, December, February, January, July, June, etc. So
we're back to the use of two digits for months.
So I date a lot of things with and ISO standard YYYY.MM.DD. Which
also sorts filenames "Stats20190701" comes after "Stats20190603".

But, to each his own, there is that "installed user base" even for
N=1 systems.




If I had such a folder I'd probably call it Jan-1-19. I also
don't sort photos by date but by topic. Naming a photo

--
pyotr filipivich
Next month's Panel: Graft - Boon or blessing?

"Mayayana" on Fri, 4 Oct 2019 21:37:34 -0400
typed in alt.windows7.general the following:

Occasionally I might have something with a number name,
but in general I don't. So in general, if I have something
I want at the top of the folder, I can just name it AA*.

And that is why I'll have a filenames _1AA*.

When I download pictures from the camera, or transfer scanned
images, a dated file folder is "useful" for some values of useful. But
I do intend to organize them by subject matter, and rename the files
themselves. Someday.

I actually do name my website server logs. Tue10-1.
Wed10-2. Etc. Those then go into a folder named
stats Oct 2019. I guess I do that because the date
is always a secondary qualifier. The main point is that
the folder contains web server logs. Stats.

Yep.
_A_ problem with naming files "Jan1" is that then next two files
are Jan11 and Jan15, not Jan2 and Jan3, And of course, the months
sort April, August, December, February, January, July, June, etc. So
we're back to the use of two digits for months.

--
pyotr filipivich
Next month's Panel: Graft - Boon or blessing?

"J. P. Gilliver (John)" wrote

| Everything doesn't actually _do_ searches: what it actually does when
| you type in what you're looking for is filter. It has already _done_ the
| search, and found everything

I understood that. To my mind it's wasteful. If I
can find a file in less than a second I don't need a program
storing and updating a database to do the same thing.


| (-: [My C: partition (this is W7) has 40.1G used, and that's almost
| entirely Windows itself, installed software, and the software's own
| housekeeping (configuration files etc.); I never _save_ anything to C:.]

A lot of that is bloat junk. A new install is typically
7-9 GB. I make disk images for that reason. Win7
will grab a copy of any old thing it comes across and
stash it in winsxs. I haven't even tried to track down
all the wasteful bloat of Win7. But I've heard people
complain about 80+ GB. So I keep disk images in case
it gets too big, so I can just dump it and start over.

| What I in my previous post referred to as librarian's dilemma is
| particularly prevalent with photos. Let's say you have a folder for the
| Smith family, and one for Fido Jones your dog; where do you file a photo
| of Fido with the Smiths? You can't make Fido a sub-folder of Smiths if
| he isn't part of that family. A lot of photo software gets round this by
| making "albums", and any single photo may be in more than one album; it
| doesn't (except for the crassest of such software) create multiple
| copies, it does it by using "tags", which are in the "albums" which are
| lists maintained by the software. I don't like this because I like to
| know where my files really are, and don't use such software; however, I
| do acknowledge the problem. (Another problem with the album approach is
| that the list formats are probably proprietary to each software, and
| maybe even to a limited range of versions of that software.)

And the biggest problem: You need to do something like add
EXIF tags so that the software knows the photo is of Fido Jones.
If a person knows how to use a file system they can do much
better organizing with a lot less work.

I avoid the "librarian" problem by not having a dog. But I'm
also not so finicky about my system. I'm not trying for librarian
accuracy. I want to be able to find the photo later. That's all.
So in your example I might have a folder named Personal with
a subfolder named Family. That would cover the dog and the
neighbors. I also don't take a lot of photos. And of the ones I
do take, I go through them before storing them. I don't just
download the whole camera storage into a folder.

| Playing devil's advocate here (I'm more like you), I could argue that
| that's still "using their computer". (I could also say that it's a
| matter of degree: _you_ [and I] are happy with the "conceptual model"
| the OS presents us with of our files and folders, rather than keeping
| track of where the individual data clusters are on the "drive".)
|

It is using their computer, of course, but they're not
using the tool itself. They're using a series of wrappers.
The file system was created as an abstraction layer to
store and access data. To use a file organizer that shows
you a folder containing all JPGs on the system that have
"Fido Jones" in the EXIF data is a further abstraction. It's not
understanding the basic end-user functionaly of the computer.

| When you ask these people where their tax records are they
| say, "I don't know but Word knows". Their photos of last
| Summer at the lake? Who knows?! "Photoshop knows where
| they are. That's all that matters.". Backup? Who knows.
| "Aconite handles that." Or, increasingly, it's
|
| I was going to say (playing DA again) why should it actually _matter_
| that they don't know where the files "are", any more than it matters
| what the chemical composition of the fuel you buy to put in your car
| (other than whether it's Diesel or "gas")?

That's not an equivalent analogy. An equivalent would
be that they don't know how to fill the gas tank, but
that's OK because their grandson is always with them
and he knows. It's a matter of not actually knowing how
to use the tool.

| but then you mentioned Aconite (which I presume is some sort
| of backup software); if that works with Word, Photoshop etcetera and
| actually does backup properly, then perhaps they really _don't_ need to
| know where their files really "are".
|
Aconite is popular with "tech support" people. someone
pays a tech support person, who in turns makes them
pay for an Aconite subscription and sets them up with a
gmail account. Aconite syncs to cloud storage. So if the
computer has problems, the tech support person can just
refresh it with Aconite. and since the person has been set
up with things like gmail, they didn't have any local files
to lose, anyway. Very convenient for tech support. But
the person with the computer is paying a lot in terms of
money and privacy for the luxury of not understanding how
to use their computer.

It's similar with Adobe subscription. People are paying
through the nose for creative suite, which is now only
available by subscription. If you don't understand the file
system and don't make local copies of photos then you
won't know that your photos are *only*
online at Adobe's site. If you end your subscription you
lose your photos. That kind of scam is feasible precisely
because people don't understand their computers and
can't be bothered. Tech companies take advantage of that.

There was a recent discussion about whether Microsoft was
eliminating local accounts in Win10. I don't know the upshot
of that but I expect that's on its way. They're training people
to use an adware/spyware consumer services kiosk. And most
people prefer that because it's easy. Most young people
don't even understand the idea of owning their data. They've
grown up with the likes of Facebook.

On 05/10/2019 15:13:46, J. P. Gilliver (John) wrote:
In message , Mayayana
writes:
"mick" wrote

| Interesting. I guess it never occurred to me to name
| a file or folder starting with a number. I can't imagine why I would
| want to.
|
| I do it all the time both for files and folders when sorting photos.
| e.g. yearly folder named 2019, then sub folders 01 January, 02
| February. 03 March, etc.
|

But that's backward, unless you live in Europe. If

No, the US way is backward, or at least illogical (-:

Yes to both. They will eventually catch up and adopt the metric system
one day. :-)
I was taught the english system at school, but once I started work in
engineering and began to use the metric system is was so obvious it was
a more rational system to use. So easy to learn as well. 50 plus
years on I still occasionally compare imperial weights and measures,
but metric is always first choice.

I had such a folder I'd probably call it Jan-1-19. I also

I don't think Mick meant he had a folder for 01 January, just that he had one
for January, which he _named_ 01 January to make it (and the other
similarly-named month folders) appear in order. (The names of the months
aren't in alphabetical order.)

Yes, that is what I meant.

don't sort photos by date but by topic. Naming a photo
010120191 is not any better than the camera naming it
P10533492.

Certainly no _worse_, though, and I'd submit it is better in that it conveys
more information than the P format does. (Though in that format, with the
digits all run together, I'd have to know the format before I could discern
the information.)

So I'd never have a folder named Jan-1-19.

I _do_ have folders named something like 2019\10\5, though not in my images
area.

All my image file names are named by 'year - number'
e.g. this year they start at 2019 - 00001
as of today the last image filed is 2019 - 8861

Categorising, naming, tagging, sorting, keywords or whatever is all
done in ACDsee Ultimate Pro.

--
mick