Forum Posts Feed
WinCatalog
< Next Topic | Back to topic list | Previous Topic >
Posted by Wayne K
Jan 27, 2014 at 08:11 PM
22111,
When I said I wasn’t too worried about proprietary I was referring specifically to file managers, not PIM’s. I see it as less risk because we’re talking about the possible loss of attributes attached to files, not original notes that you’ve created.
Regarding Xplorer2, I have the pro version and did mention it at the beginning. You can’t modify columns.
All this is related to the previous thread about being able to change file attributes so they file information is portable and available no matter what file manager you’re using. That would be nice. I tried creating a poor man’s version of this be incorporating key info in the name. For example, I would include the author, source, and date in the file name for an article. It’s an awkward system because the file names get so long. You also can’t sort files by attributes but you can kind of work around this by doing key word searches instead.
Donald has confirmed to me that he plans to add more customizable columns to XYplorer. I’ll report back when that happens.
Thanks for the other software suggestions.
Wayne
Posted by Wayne K
Jan 27, 2014 at 08:21 PM
Based on what I’ve seen so far, I’ve concluded that its better to stick with mainstream file managers based on the following.
1) Even if you find software that offers better handling of attributes/columns, if you try to use it to manage your files you’ll immediately start missing all the other features that the top-line file managers offer (powerful formatting, filtering, etc).
2) To be workable, the other software has to take over all file management. It’s too much trouble to be constantly transferring files from the primary file manager to the alternate program, and keep it all up to date and synchronized.
Wayne
Posted by 22111
Jan 27, 2014 at 10:21 PM
I
Jim and Wayne,
it occurs to me that since we’re speaking defunct functionality, even though you say it works (I have XP only and so cannot check), you rely upon that functionality that is inherent to the file manager in question - please try to read that same data, from X2 to Explorer, to DO… this will be rather instructive.
Jim,
Yes, I’m speaking of those Alternative Data Streams, some months ago I spent 2 days with the problem, and indeed, they have officially been shelved by MS - it’s all the more of interest how MS does it now with Explorer, and as Wayne says, we need some system where the same file info is available whatever file manager we use - hence the interest of ADS (on ntfs at least), but which has been dumped, without having being replaced.
Wayne,
Your last post says it all, the functionality would have had to be within a (good, paid, ok if it’s expensive) file manager, but it is not (anymore) - did I say it would be interesting to show if at least the X2 colums are 100 p.c. compatible with the DO ones, and vice versa? (Since both have become sort of “proprietary” now, in 7/8 (and beginning with Vista if I remember well).) - All those additional programs add too much fuss, especially if they do not synch themselves.
Of course we are mixing up two paradigms here: Virtual folders (= distribution into groups from start on), and “column” data (which is for filtering on that afterwards), but as long as we’re aware of our discussing two concepts at the same time (since they fulfill the same task)...
Let’s speak again of the “individual” columns you think you need. As I had understood it, most of them are standard attributes, not of the file manager in question, but of the ADS / ntfs system, i.e. “author”, “last changed”, etc., etc. - these were perfectly “exportable” between any file manager, in ADS days (up to XP).
On top of these, your local Win version seems to add the whole bunch anew, i.e. you have got all these in English, and then even perhaps in German, French, Spanish… So there must indeed be a way to add “individual” columns to ADS, since the local Win version does exactly that. (But then, there would be “exportability” issues, not from one file manager to another, but from one Win installation to another in another language, and both should be XP.)
I would have tried to find more precise info on these matters, had it not been the fact that anyway, that system is doomed. I think I remember that a third-party add-on does that in Explorer, even in 7/8, you always can read (but not write?!) ADS data, but you’d need X2 or DO to write it into those ADS of these files.
Again, I tried to discuss these matters with X2 developer, but he fervently holds to any know-how advantage he might hold.
Also, at present, this ADS is always there, but not “acknowledged” by MS anymore; in “tomorrow” ‘s file system, it might not even be available anymore, even for such proprietary solutions offered now by X2 and DO - which arises the question where XY does store its metadata, and why it has been called “proprietary” to begin with? I think I remember it’s stored in a db over there?
III
Wayne, with all reservations re ADS today, why do you need MORE “columns” (ok, X2 calls them that, but they are ADS attributes) than are available in X2? I have these attributes in 2 languages, but I counted them, they are 67! Let’s say in your English-ownly system, these are about 35 only, and you cannot enter text data in a “last accessed” attribute, that’s accepted, but there are numerous such attributes, and in the “Comments” attribute at least, you can put multiple codes, keywords, whatever! Would’nt that be enough, with carefull planning? (Which is necessary because it’s very cumbersome to batch-change all this, afterwards.)
II
Wayne, trying to get too much info into file names is pure horror, as you have seen for yourself. That’s why invented my multiple folders system where the respective folder is displayed for every outline item, and this means, instead of doing metadata by ADS anymore (or within file names), I try to put every external file into the proper sub-folder.
Then (since this makes only for “sorting by ONE criterion), for author names and such, I use “Search Everything” (SE), which only searches within the global “container” folder that “holds” these detailed folders, i.e. I try to not code info that is readily available by SE anyway (but I very often add additional, explicative info to the original file name, in the form “original file name (my additional info).suffix” (in order to not have to open pdf’s too often such for a “what was it again upon?!” afterwards).
Then, there is file linking; in a system like mine, with hundreds of folders, file linking is very efficient. Here, I have to acknowledge that my advise isn’t as good as my reality: My folders are named in the form “cio - IM - Outliners” = I do NOT describe the first code, C = Computer; I’ve got just 30 such codes, and I memorize them. Then, for the second code (= often many, for each first code, = may be over hundred of them, no need to memorize them), I do a SHORT explanation, here “IM” for “i”. Then, the third code, as mnemonic as possible (so some details have weird names), “o” for “Outliners”, this I write in “full description”.
Now the trick in my system is that for any file operation, it’s just the code part that I need, and the same goes for folders. I not only have a folder “cio - IM - Outliners”, but whenever I want to switch to that folder, I just do an “F7 cio”, and AHK does not just enter the “enter” (I’ve got 1 sec and a half to enter the 3 characters, so I better decide upon them before pressing F7), but also replaces, internally, the “cio” by the full folder name.
This means, whenever I have got some OTHER file, anywhere, which I also would like to be able to access from within the cio folder, I simply select that file in ANY file manager, then press “control-F7 cio”, and I don’t even have to check there: In any case, that file, with its full file name, will have got a link file within the “cio - ...” folder (and if really there is a technical problem, I’d get a message).
Now, if I say, do similar, I know that in a regular file system, this is not “work” for (literally) 2 seconds, but a lot of fuss, and this prevents pc users from using such a system, hence the interest of extensive tagging instead, but this tagging, too, should be done with codes; then, either anything in just one big folder, and filtering, or in different folders, and SE (or in X2 and some other file managers, even filtering “over” many subfolders at the same time).
What about a non-intrusive “code” symbol like the degree symbol, which probably you’ll never use in any file name (internal or external)? Then, “°cio - original filename”, “°cio - another original filename (and some short comment of yours)”, “°cek - original filename.pdf” for e.g. some intro into “Computer - Editors - KEdit”, etc., etc.
Why the degree sign? Because most “searchers”, like SE, would produce many false “hits” when you’d search for such codes, most of which are ubiquitous as parts of words, since it’s not so simple (or even impossible) to search with these special tools just for characters at the beginning of a filename.
Then, the same principle applies to other codes, ToDo/Delegation/Belongs-to-project-xyz codes and such: Here again, you need some code symbol you otherwise will never use (but there are plenty of them available, for standard fonts like Arial, Tahoma and such, between abc and the accented chars, or then, the ° again, and next char being a char you will not use as FIRST char of any naming code), and then hold it as simple and short as possible, at the BEGINNING of file names.
Example for an external file: “°cek - Some Intro into KEdit”, which is within your general folder (or within your folder “sw” when you prefer trans-folder filtering). You need this within a certain project, let’s say you have some qualitative analysis work to do, let’s say a project called “slf - Sociology - Film Buffs in London (from Art House to illegal file exchange (whatever)”, and for which you need KEdit, and you want to have this intro handy there. Now I would simply make a link from the original file in cek that appeared in slf, but you will have just a slf folder, too, or in the folder “university”, “writings” or such, or then, your “all in” folder.
Anyway, you need cek… within your slf project. So you would RENAME your original file, from “°cek - blahblah” to “°cek - (°slf) blahblah” - or to “°cek - (°slf) (°pks) blahblah” if you need it in another project, too. Now you want your kid to have a look into it, for its homework, and you rename it “°cek - °zt (°slf) (°pks) blahblah”, or, instead of °zt, #t; both are for “Tom should have a look into it” (let’s say you have 2 children, Tom and Jerry, that leaves you anything else but zt and zj available for other coding (whilst the # variant would be perfect if you had two dozen collaborators, leaving the z for other, various codings).
Now in your project slf, you have SE search for °slf (the ° being placed there by macro), or you simply FILTER FOR °slf, and your file manager shows you any file that you need for your London ex-cinema-goers project.
Here, one very important detail: Don’t despice free file managers: I own 6 or 7 paid ones, but FreeCommander XE is a wonderful piece of free sw which does better filtering than most highly-paid ones.
I know all this is not beautiful, because of the multiple add-ons to file names, but ADS isn’t available anymore, or just within one single file commander. My system (references to files in multiple contexts/folders) seems to be best… when I’ll have overcome the prob that all those fine quick links are broken whenever I rename the original file name… It seems that beyond XP, and with other formats than just .lnk files, there are possibilities. Remember I do NOT have multiple context coding in my system, just ToDo coding (and even that could be replaced by “virtual” folders, i.e. real folders, named “Today”, “This Week”, etc., just containing .lnk files pointing to the original files - remember, one such link / “tag”, in my system, takes 2 seconds, incl. the entry of the target name(‘s 2 or 3 character code).
I’m more than willing to discuss other file naming conventions, but for a start, the above details might justify some batch renaming, then seeing how you can do it best: Select the files that “go” into one project / context / subject / whatever, add the common code, do it again for other subgroups… do not try to do it all manually. And for that batch renaming. Above, I said I often add (comments).suffix after the original filename; here for multiple context, etc. coding, when you do batch rename, you will always replace the “) ” by “) SomeAddition”, so try to rename/code the most important file groups first, then add secondary group assignments… and depending on your batch renamer (lots of them are free, and some of the best ones are in paid file commanders), you might even avoid separating different codings by spaces, since this would bring the possibility to not only add at the end of the coding section of the file names, but also at the beginning - of course, you can use brackets and braces, but they are even uglier. Or then, separate them with semicoli, and leave out the parentheses altogether - that’s probably best. (But don’t leave out the °°° - you’ll also need them for any renaming, afterwards.) (Well, thinking about it… why any (); or whatever, anyway?
Have it this way, imo:
°xpc °ats °flz °za °zt - Original Filename (Author or such).suffix
And you’ll always be able to replace the ” - ” by another ” °plt - ” then. But code your contexts, or else you’ll get file names 30 cm long and will lose all faith in pckind. (And yes, I would have preferred the preservation of ADS comments; btw, X2 even lets you filter by them, as long as it goes.)
Posted by Wayne K
Jan 28, 2014 at 12:19 AM
22111,
I was excited when I saw the many columns available in Xplorer2 but you can’t modify any of them except the comments attribute, so they’re worthless. At least I don’t think you can. I didn’t systematically try every one of them but I tried a at least 6 or 7 and could see I was getting nowhere. I then contacted Tech Support. They didn’t dispute my observation that you couldn’t enter information into their columns. They also didn’t seem to think that was a shortcoming in their program or something that should be addressed in any way other than to say “you can’t do that”.
As I said above, the big take-away I’ve gotten out of this exercise is the conclusion that any two-step solution that involves exporting files from the primary file manager to some other program is not a good solution. The solution has to be within your primary file manager. That alone simplifies the issue enormously.
I’m sure there are other ways to accomplish what I’m trying to do, and you’ve suggested some for me to think about.
I’ll reply to some of your other points when I have more time.
Wayne
Posted by 22111
Jan 28, 2014 at 03:33 PM
Wayne,
“As I said above, the big take-away I’ve gotten out of this exercise is the conclusion that any two-step solution that involves exporting files from the primary file manager to some other program is not a good solution. The solution has to be within your primary file manager. That alone simplifies the issue enormously.”
As said, we convene here by 100 p.c.
I
As for ADS attributes in X2, here again, I never said your info is wrong, and I fully acknowledge all that “other” ADS attributes there cannot be “changed”, i.e. neither renamed nor changed in their respective information “format”. So, from your insistance on this point, I take that I should have been more clearly in what my suggestions meant: In fact (but without ever wording this), I constantly spoke of “abusing” some of those not-needed attributes, wherever their data format permitted this, and from what I saw (several months ago), some of them lend themselves perfectly to such “abuse”.
This being said, the respective attributes/columns cannot be renamed, and it’s quite awful to have relevant info in attributes with misleading names - it’s all about ugly, provisional solutions, and with both X2 and DO developers NOT responsive on this matter (which could indicate that they consider this subject, too, but don’t want competitors to get any info on the possible solutions they might have in mind, for future Win versions - it also seems there is a lot of technical know-how here about HOW to address/process ADS data… let’s remember XY never used this format, in spite of constant user demand for it).
And again, MS has dumped ADS, dumped in the sense that just in current Win versions, it seems to be there yet (since X2, DO, etc. can address/process it), but they can take it away anytime (well, it does not seem to be a feature of Win, but of ntfs, which is certainly the only reason it’s always there: MS WANT to dump it, but they cannot, without dumping ntfs at the same time, and they don’t dare to do so yet ;-) ).
II
That’s (and the absence of real info about possible alternatives after searching for days in the web) why I think a “file cloning system” is best (= a more elegant “virtual folders system” since it only relies on folders being really there (whilst you can NOT “see” those “file collections” / “virtual folders” created by X2, DO, SC, etc. within the regular folder trees, even from within the same file commander you will have created those, let alone looking into your file system with any “competitor”), coupled with some additional info within the file names (= additional “tagging”); the multiple “contexts” here being shown just from any file manager (since you’ve got any file (= reference) in any such context (= real file folder) where it might belong (, too), and any “additional” info then is available by filtering for codes or text (authors and such info) within the file names.
I’m aware that these “text tags within the file names” cannot be combined, for such filtering, in current file managers (or did I overlook such functionality anywhere)? This means, you can filter for term1, for term2, and even for term1 OR term2 in many of them (you must just check by which syntax, e.g. “term1;term2”, or just by a space between them, BUT this will only add anything containing term 2 to anything containing term1, NOT just show files containing both terms, and which would be needed very often.
Hence the need for additional search tools, for such tasks, and here again, we’ve left our “do it all in just one single file manager” objective. That’s why I would like to have a combination of file manager AND outliner (as described in other threads), and which would have more or less “full file M” functionality (which is not the case for UR, e.g., and which does “more” than most of its competitors do, for the moment).
This being said, many “additional” functionality seemingly “within” current file managers simply run additional tools in a window on top of the file manager’s window, then that additional window is automatically closed again by the file manager, and both the trigger and the close/hide again function could be realized by a macro, for another (but fully-functional) provisional solution.
And to say a last word on “tags-in-the-filenames”: I’m aware that my “virtual folders” system above is just as worthwile as is cloning in outliners, and that the same limits apply: It’s just the “primal tag”, different (= “I need it here, too”) context - it would be aberrant to create folders (or, in an outliner, additional headings) for “combinations” (e.g. “I need it for project abc, AND the author is xyz”; but there are SOME cases where this CAN be done BUT as a subfolder: folder “project abc”, and then, if you’ve got 20 papers from just one author you need there, why not have a subfolder “author xyz” within your folder “project abc”?)
In general, though, some combination of “primal tag” (whether you realize that as a “virtual folder” or by a tag yet), and then, secondary tags (mostly by real tagging, be them in the file name, in an ADS or within the file format itself (pics, MS Office…)) is the way of doing things, hence the interest, if you do it by info in the file name,
of coding (for briefness, not only in display, but also for then filtering/searching for these tags!),
and of standardization (as such info is standardized e.g. for pics, and in ADS attributes, again both for visual reasons, if in the file names, and for easy and safe/complete access to the relevant files afterwards (i.e. you introduce variants, you’ll inadvertantly leave some of the possible “hits” out, or the other way round, you just want xyz as an author’s name, but you’ll also get files where the same term is part of the original file title (e.g. “Answer on abc’s Paper “def” in the “ghi Journal” in October 2013” - this will leave you not just one, but FOUR chances for getting false “hits”))).