always resolve junctions?

[nikos]

I'm having problems with junctions for the new version, and I'm thinking I should just translate them all the time.

Say you have a folder junction under c:\temp (called junk1) which points to d:\other. With the current version, browsing junc1 would enter a (pseudo)folder called c:\temp\junk1 (which would actually be d:\other).

Would anyone have a problem if I always made xploer2 to browse the real target folder? I bet most people don't even understand what i'm on about

[Kilmatead]

Your blithe condescension to a power-user-base notwithstanding (the dirty scallywags), grim reality is always preferable to pink elephants.  I vote for a strict literal usage - why bother with pretence?  (Or, at my age, foreplay...)

To be honest, I hadn't even realised you wasted your time making the titlebars into pseudo-folders in the first place.  That said, I could see some people depending upon the link-name as part of their mental structuring process during workflow, so I'd be surprised if there were no dissenters.

[FrizzleFry]

What is the advantage of the current behavior?

[nikos]

What is the advantage of the current behavior?

i can't think of any either way. As kilmatead says it may only annoy some people that use junctions to make things tidy

[fgagnon]

personally, I hate the use of pseudo-folders, much preferring to know where I am in the 'real' folder structure. Having said that, I can force myself to appreciate why some may like them functionally.

[FrizzleFry]

Say you have a folder junction under c:\temp (called junk1) which points to d:\other. With the current version, browsing junc1 would enter a (pseudo)folder called c:\temp\junk1 (which would actually be d:\other).

Would anyone have a problem if I always made xploer2 to browse the real target folder? I bet most people don't even understand what i'm on about

So the new behavior would be to show

D:\other

rather than

C:\temp\junk1

when you browse into junk1?

I guess it would not bother me but it would be acting differently than WE.

[nikos]

what i've discovered is that junctions to network folders don't work in the "pseudo" mode, so it's easier to just resolve all of them

[FrizzleFry]

Do you mean symbolic links? I thought junctions didn't work on network folders.

[RightPaddock]

That said, I could see some people depending upon the link-name as part of their mental structuring process during workflow, so I'd be surprised if there were no dissenters.

Precisely but...  

it goes beyond the mental structuring process during workflow, it goes to your actual operating process during workflow!!

My main use of junctions is provide access to the same data sets in different contexts (not ONE different context, up to SIX different contexts).  The folder in which I initially store the data sets whilst consistent is in fact arbitrary - it happens to be Collection Date, but it could have been Location, Manufacturer, Measurement Type etc

An essential feature of the current scheme is that I can see where I've come from, and use backspace to get back there - usually to look at another data set in the Same Context.  If you take that away then you'll give junctions the same behaviour as a shortcut - backspace would always take me to the Collection Date!  I am not denigrating shortcuts - I use them by the truckload for Context Switching - in the opposite pane via Alt/Enter.

Are you planning to impose the same non conformist behaviour on folder symlinks?

rp

[Kilmatead]

An essential feature of the current scheme is that I can see where I've come from...

Congratulations, it's not often that someone can change my vote by using 16 simple words.  Well done.

Nikos, he's made a point which goes beyond just "people that use junctions to make things tidy".  Add that to FrizzleFry's observation that it would be acting differently to WE, and the "easy way out" may not be so easy.  While one may presume that <Alt+Left> would still behave as expected (in league with how a <Ctrl+L> resolution can be reversed now), the context of the junction itself would be lost - and that's not a good thing to lose.

Since <Ctrl+L> resolves junctions "literally" now (it does not invoke the pseudo-folder), there's already a precedent for making junctioned-networks special-cases and leaving the rest as they are now.

[Thracx]

...Would anyone have a problem if I always made xploer2 to browse the real target folder?...

Yes, it would annoy me greatly.  However, I believe X² already does this in some cases but not most (as discussed here).  While it would be to my bane, X² would be a better app if it was always self-consistent.  I say make the change so all SymLinks are fully resolved.

Alternatively, can't you just track the real-location as separate from the user-location (shown in the Address Bar)?  It would then be easy to always use the 'real' directory location in your code, but display that user's perceived location in the address bar.  It would then be trivial to provide an option to enable/disable preserving the different path - then you and fgagnon can have the address bar always show the resolved path and I can see my virtual paths.

...While one may presume that <Alt+Left> would still behave as expected (in league with how a <Ctrl+L> resolution can be reversed now), the context of the junction itself would be lost - and that's not a good thing to lose...

And you see the light - thanks RightPaddock!!!! :=)

what i've discovered is that junctions to network folders don't work in the "pseudo" mode, so it's easier to just resolve all of them

How so?


PS.  Be sure to test with NTFS volume mount points!

[Kilmatead]

Umm.... just a thought: wouldn't resolving all junctions destroy the possible integrity of the /Rootfolder command?  

Edit: On second thought - maybe it doesn't.  <Ctrl+L> blocks direct transfer to the target, so only the target's contents would be available (such as from a link to another drive)... but it would sure be confusing as heck if anyone looked at the addressbar or pane header.

[nikos]

junctions aren't resolved, only network ones

[Kilmatead]

You know, it would be nice if you remembered to tell everyone when you come to a conclusion about something!  You did ask the question in the first place.  Or was this just another one of your famous "Osmosis Tests" just to see if we're more clever than your average bunch of ragamuffin-children?