topbanner_forum
  *

avatar image

Welcome, Guest. Please login or register.
Did you miss your activation email?

Login with username, password and session length
  • Thursday November 6, 2025, 1:57 am
  • Proudly celebrating 15+ years online.
  • Donate now to become a lifetime supporting member of the site and get a non-expiring license key for all of our programs.
  • donate

Recent Posts

Pages: prev1 ... 11 12 13 14 15 [16] 17 18 19 20 21 ... 109next
376
Thanks for sharing that. I think it's good stuff.

Some complementary thoughts:

- Multiple ways to present the software repository organization : Couldn't all software titles be put in a DB and then, through different filters/queries and sort methods, we could see just what we want in a table/grid like interface ? I'm thinking of Something close to Mantis in terms of functionality, but maybe simpler and nicer (aesthetically).

This would be integrated within the site. It would be easy for the admins to see exactly what needs to be updated, when it was last updated , etc. But maybe is this a terrible idea...  :)

(This wouldn't have to be the only interface : DB data could be formatted differently in other pages where organization would be more static (standard web page))


- Redundant info problem (especially regarding bug tracking and suggestions, I suppose...) : as far as bugs and features suggestions are concerned, I think that all that is needed is a relatively strict way to do things. But since it's  hard to force users to adhere to specific (even simple) rules, it needs to be mostly implemented by the admin's/mod's responsibilities. This doesn't mean that the users shouldn't adhere to a couple rules (e.g. : if you want your bug to be discussed, post in the bug section and tag it as a bug...)

In the InfoQube forum, we've implemented a simple way using the forum and Mantis.

Mantis is the developer's turf, but it can be browsed and issues can be added by anyone there. In reality, only a few users do take the time to do so and it's not a problem.

Anyway, the procedure goes like this :

1- Everything is discussed in the forum.

2- When something is "confirmed" (issue, important feature...), a- it's entered in Mantis with the thread's reference (hyperlink). Mantis isn't a place to discuss things (some discussions take place there, but they are extremely limited and focused)
b- a code is entered in the forum's threads title (so that's obvious that issues have been filed), c-the mantis issue ID is entered in the last post of the thread. (Obviously the mantis Id could just be added to the forums title thread too, so it would be a 2 steps process : a, b)

It looks complicated but it's not. it just means going through all bug reports and suggestions once a week (probably not much more than 10-20), and copy paste info from the forum to Mantis. Some of it could even be scripted (steps a, b). Once that is done, one can forget about the forum, and maybe just update the related threads with info when a serious bug has been fixed. But even that is generally unnecessary.


I think redundancy is probably not as bad as you make it sound.  :) E.g. : as you know accounting systems are based on efficient redundancy, and so is data security, etc. We just need a way to control it so that it doesn't become chaotic. Redundant material used as a "history" for verification purposes needs to be identified as such and then... forget about it, "archive" it (i.e. : let it sink in the forum's abyss).

No idea if this has any value... Sorry if I'm repeating the obvious or wasting your time.  :)
377
General Software Discussion / Re: DVCS ?
« Last post by Armando on March 04, 2011, 10:07 PM »
I've settled on Git for moving fSekrit forwards, but that's mainly because of Git's "blow off your legs" power with regards to history rewriting - I want to make the codebase public, but retain no memory in the public codebase from the non-public versions. Yet, at the same time, I want the private repository to have FULL history, and do that without committing future versions to both public and private repositories. Git supports that (even if I haven't worked out the details of how to set it up 100% :)).

One question : You'll still be able to commit code from the private to the public repo even if you squashed the changesets/history in the public one ??? I most probably don't understand properly as I'm not sure how that would work.

In any case, I've edited Mercurial's ini file to add the "collapse" extension and it seems to load properly. Haven't properly tested it yet. Collapsing changesets is probably not too dangerous when working alone, but I don't know about team work. Potential for disaster.

History of a single function that moves? Interesting. That's not something inherent in how git stores the version history, though, unless I've missed something - Git stores the full version of each file for each commit, whereas both subversion and Mercurial store changesets (iirc Mercurial also stores a full-version "every once in a while" so moving between versions doesn't become too slow). But IOW, if Git can do it, it's gotta be history analyzing magic smartness that could potentially be done by everybody else too?

That's what I read on stackoverflow. Don't take my word for it! I'm going to try to find out more info on that.

From what I've seen from various comparisons, Git and Mercurial offer most of the same features, as long as you don't need the über-wicked geek functionality (Mercurial pretty much won't let you blow your legs off). Also seems like there's not massive speed differences for "normal" size projects.

I've seen a number of benchmarks. AFAICT, it varies depending on the task, OS, etc. so it's hard to say. In any case, I'd say that when working locally, we're talking about less than a second. Bazaar has some nice benchmarks numbers on its page, but... I haven't found the same results anywhere else.

Git seems a bit more down and dirty, and the tortoise version still isn't as polished as the tortoise versions of other version control systems... it still shows a bit that Git was originally a whole bunch of shell scripts, instead of designing it "properly" as a C/C++ library with a "proper" front-end :)

I just looked at smartGit and it looks really nice. I'll see if I can play with it during the weekend.


I probably shouldn't but I'm currently looking at Git.
Oh yes, you should, if you have the time, resources and interest in it.
Making a more educated choice now (early/pre-start) is way better than to change half way through your project, if you then discover a feature in the other solution you absolutely must have, for whatever reason.

Thanks for the encouragement. I just have to be careful to not spend too much time obsessing about details -- something I'm pretty good at.  :)  But you're probably right. That's why I'll check out the Git option bit more carefully.


I use git with great success, but it does have drawbacks, namely poor support for Windows platform.

It is very fast even with large repositories.

AFAIK mercurial has much wider platform support than git, but is also rather slow. This is from hearsay though.

I chose git at that time because it was the new shit.

Eheh...
Git effectively seems a bit speedier. But, concretely, I'm not sure if I'd be able to see the difference in my everyday work.

The 'portable Git preview' software (for Windows) you can get from softpedia looks to be very promising.


And together with SmartGit (free for personal use) it starts to look quite nice as well.

One hell of lot better than WinCVS at least  ;)

**EDIT
Added the SmartGit part.

Thanks Shades. I downloaded both. Will report back.
378
General Software Discussion / Re: DVCS ?
« Last post by Armando on March 04, 2011, 03:19 PM »
Hmmmm. I probably shouldn't but I'm currently looking at Git. Who's working with Git here ? Are there reasons why you chose it over Mercurial or Bazaar ?

(Moving a function from one file to another and Git can tell you the history of that single function across the move... ? Really ? Sounds appealing. I guess this is where content tracking (instead of file tracking) shines.)
379
General Software Discussion / Re: DVCS ?
« Last post by Armando on March 04, 2011, 09:42 AM »
Thanks.
I read a few things about overlays,  but nothing seemed to really apply. I turned ON the "Local Disks Only" setting and see if it helps. (Explorer extension settings --> icons)
380
SimplyFile sounds nice...  :)
381
General Software Discussion / Re: DVCS ?
« Last post by Armando on March 03, 2011, 08:18 PM »
Nope, some language/tools/developers using vcs controlled source induce so many changes, that can be relatively unimportant, but need to be checked in at one moment, while in another environment, a single line changed can make a world of difference that needs to be committed at once. So change-counting doesn't count :P

Sure, in a way change-counting doesn't... count. But in my relative and very imperfect world, it does.

And why couldn't one adjust the parameters depending on the project ? (A simple backup comparison: some backups need to be real time or performed every few min, but others can be performed monthly.)

I understand perfectly well why one would never want to commit randomly. However, I'd vastly prefer to be slightly annoyed by a pop-up from time to time, than potentially loose several hours of work, or simply find myself committing a huge chunk of code that should've been split in smaller part... because I forgot to commit my work.  :)

Of course I could then rely on autover, AJC or File Hamster. But... Far from ideal.

---

On another note : I wish I could rely on the icon overlay... But they are slightly unreliable here. They change only on F5, and sometimes they don't. I have to open/close the repo/folder for them to update.
382
General Software Discussion / Re: Solving Issues - Pain and More Pain
« Last post by Armando on March 03, 2011, 08:01 PM »
 :) That type of thing happened to me a lot lately, and it involved XML (and character encoding).

Will focus on improving my problem solving algorithms...
383
Site/Forum Features / Re: Shortcomings of DC and How to Improve
« Last post by Armando on March 03, 2011, 03:55 PM »
The "home" page might be important... However, maybe it's just me, but I very rarely see/use it.
I'm mostly in the forum, and I navigate from there... My home page "is" the forum -- especially since I can access everything from the forum. I wonder about others...

the idea of changing the home page is for new visitors. I think it would be good to make it more dynamic, or at least to lead people more towards the software, and the forum

I think it took me a few weeks to realise there was even a "Home" page here, when I first registered ! But sure, I understand that point of view. A home/Start page is important. But only if really gives the user something important that can't be found on other pages.

A question : do most visitors enter DC by the front door...? As we all know a website isn't like a house. You can usually "visit it" from many different pages, especially when arriving from a search engine.
384
Site/Forum Features / Re: Shortcomings of DC and How to Improve
« Last post by Armando on March 03, 2011, 02:07 PM »
The "home" page might be important... However, maybe it's just me, but I very rarely see/use it.
I'm mostly in the forum, and I navigate from there... My home page "is" the forum -- especially since I can access everything from the forum. I wonder about others...
385
General Software Discussion / Re: DVCS ?
« Last post by Armando on March 03, 2011, 01:45 PM »
In relation to what f0dder said...

First, there's no way to really tell when a file modification is done and complete - the best heuristic you could go by is "file has been modified, and now hasn't been additionally modified for n seconds".

Second, you don't want commits based on random modifications - at least when programming, you'll often be doing a related set of changes across a few files, and then commit that. Think of it as not commiting files, but committing features :)

"Auto-committing" isn't the way to go... But an option/feature in Tortoise to just popup a reminder after n file modifications in a specific timeframe would definitely be useful, especially since it can't integrate with all IDE (or whatever) out there.

Wouldn't that be a sound feature request ? Would you use it ?
386
Living Room / Re: Gmail accidentally resets some accounts
« Last post by Armando on March 03, 2011, 01:33 PM »
Thanks. Will check that out...

Some stuff can be done manually :

http://manas.tungare...gle-calendar-reader/

A little script could be written to automate the process, I guess.
387
Site/Forum Features / Re: Shortcomings of DC and How to Improve
« Last post by Armando on March 03, 2011, 01:07 PM »
they should be linked back to a purpose that either (a) mouser has prioritized or (b) has been pointed out and affirmed by more than one person as a priority requirement for some tangible reason.

exactly...
388
Site/Forum Features / Re: Shortcomings of DC and How to Improve
« Last post by Armando on March 03, 2011, 09:58 AM »
I agree with Mahesh2k : a poll on a couple well thought points. If that's really needed.

Instead of just wondering how to change or modify things because a few users seem utterly dissatisfied (and I agree with app : positive comments would be welcome so that it's apparent that we're not just trolling here but actually being constructive), let's react to real needs : what the community needs. It's possible that 70% of users want a dramatic change, on multiple aspects... But it's hard to tell. It's also possible that mouser's bored to death or feels that the website etc. is a real barrier to what he needs to accomplish... and that he wants to make dramatic changes. If that's his wish... so be it.  :)

Phitsc mentioned the changes lifehacker and the like made "recently". Most of the changes there, I don't like.  :) Again, App's example of The craiglist is a fine one... Thanks app. (The Craig list works damn well... Why fix it ? Is it really worth it...). I like change in technology and forums, but not when it disrupts functionality and distracts from the content. Of course, there's the occasional formal addition which creates a need that wasn't there to start with, but that's rare...

Apart from the absence of wiki functionality (I like to use it when it's available... If that's part from the new SMF, so be it..,:) ) and a few interactive options (like the ones found on stackoverflow), I like the site. I come here to share content.

[EDIT : timns brought some good points... Of course since he speaks of wiki...  :P ]
389
General Software Discussion / Re: DVCS ?
« Last post by Armando on March 02, 2011, 10:15 PM »
You can look around, but from what I read in the last few days, it seems that VisualHG which sits on top of TortoiseHg would be the way to go. So you basically install TortoiseHg (usually bundled with the latest stable Mercurial), and then you install VisualHG.

A few links... but there are many others.
http://www.eworldui....r-Visual-Studio.aspx
http://kiln.stackexc...alhg-vs-hgsccpackage
390
General Software Discussion / Re: DVCS ?
« Last post by Armando on March 02, 2011, 11:33 AM »
That's probably the main reason the icon overlays in Windows Explorer turn red if something has changed locally. You have to take action yourself though, no auto-commit (nor would I want that).

You haven't mentioned the programming environment/IDE you are using for your project. Maybe that has an integration, as most of the big IDE's have it (NetBeans, Eclipse and descendants, Visual Studio). Some of them can even warn you if you try to close without committing your changes.


Yes, the icon overlay is a nice addition. I'll see if that's enough. I could also just put all my code in Dropbox. Currently I use the free autover.

As far as the IDE is concerned, I'm now working on a VB project. The defunctish VB6, to be precise.

So there's "nothing" for that, AFAIK. Well... there's VSS 6 but... After reading Visual SourceSafe: Microsoft's Source Destruction System and Visual SourceSafe Version Control: Unsafe at any Speed?, I chickened out (old articles, but relevant to the available VSS version...).

For Java projects I'll use Eclipse and, yes, I saw plug-ins/add-ons... But I haven't looked closely as this is not what I'm currently working with.

First, there's no way to really tell when a file modification is done and complete - the best heuristic you could go by is "file has been modified, and now hasn't been additionally modified for n seconds".

Second, you don't want commits based on random modifications - at least when programming, you'll often be doing a related set of changes across a few files, and then commit that. Think of it as not commiting files, but committing features :)

You are absolutely right : committing isn't about saving random set of changes. What I hypothetically had in mind was a separate branch for automated versioning (which could of course be merged with the "manual" branch... Not that it would have to, but could) -- e.g. each version of that branch would correspond to a "save". However, it's a non-existent solution and I can see why this would not be such a great idea (in terms of clarity, mostly). Using the available solution (i.e. : a separate software to do the arbitrary "versioning" part) seems like the best solution anyway.


One thing that git was able to do (but I'm not sure if Mercurial supports this or not) is collapsing a set of changes at the 'tip' of a branch into a single change.  Git also lets you drop changes that are at the tip (though it's pretty easy to get them back if you do it by accident, since the change isn't really gone, it's just not at the tip anymore).  What these features enable is that you can commit your local changes at will, without too much regard for whether or not they're 'correct'. Until you push the changes, they aren't seen by anyone else, and when you have something you want to push publicly, you can squash (that's the term git uses) the several changes into a single, coherent change, and push that to the public repository.

Thanks for the heads up ! I just looked and there's a mercurial extension that does exactly the same thing : CollapseExtension. And there's a lengthier solution (no extension) described there : Concatenating multiple changesets into one . Alternatively, there's also this : Rebase Project .

The collapse extension sounds useful :

Usage Patterns

This extension is particularly useful when you follow the practice of committing regularly to your local repository, possibly before you've got anything sufficiently complete to be put in a shared repository. Rather than filling the changelog with changes of the "OK it almost works now" type, you can collapse all of your intermediate commits before pushing to the shared repository.

N.B. 'hg collapse' is destructive; it rewrites history. Therefore you should never collapse any changeset that has been pushed to another repository, unless you really know what you are doing.

I'll keep digging... But I'll also eventually have to start coding.  ;D
391
Site/Forum Features / Re: Shortcomings of DC and How to Improve
« Last post by Armando on March 02, 2011, 10:28 AM »
On a practical note (not directly related to the dialogue here, but on-topic).

One thing that has bothered me over the years is -
sometimes people put a lot of effort into gathering info about a particular subject. Think of Steven Avery's list of software free for business use. Crush's and my list of Disc Catalogers. I'm sure there's others. These tend to die a death because the load of maintaining them is too much/too boring/too whatever, for one person.
I would love to see some sort of a wiki aspect/area that this kind of thing could be moved to. Re Stephen's efforts, I posted an index of sorts here. As I say in that post "It's a brilliant resource, it deserves as much promotion as it can get !!". If it were a wiki it could continually be updated and would get better and better known - instead of sinking into oblivion...

This may have been suggested before (I believe it was but only casually) and certainly doesnt fit into the forum setting, and I have no idea about the technicalities but I'm just throwing it out here again.


Re Reviews -
I can see why the reviews section (full- as opposed to mini-) died a death - too much work.
I wonder could a wiki style section be good for that too? Give it a structure - e.g. add your pros and cons.

Same as Tom here. It was probably discussed before, maybe was it even discussed 100s of times. Who knows... In any case, the wiki model for reviews and in depth discussions about certain type of software would be nice. Many threads here could serve as starting points to some of these wiki entries.
392
General Software Discussion / Re: DVCS ?
« Last post by Armando on March 02, 2011, 10:06 AM »
So, I got into Mercurial and like it. I also subscribed to a bitbucket account, etc. It definitely seems fast, unintrusive and powerful (some little "side effect features" like the  propagation of changes across file copies is quite nice).

Some early "2¢ thoughts":

While the documentation is generally superb, I found that the support concerning the actual organization of repositories not that clear (e.g. : can I create sub-repositories inside a main repository ? should I have one general repository containing all sub-projets ? or should I create a repository for each sub-project and nothing for the encompassing project folder, etc.?). Anyway : I just created a repository for each (sub) project since I didn't find any simple way to manage repositories hierarchies with TortoiseHg. (Speaking of TortoiseHg, I first started learning Mercurial by reading the... Mercurial manual. And I must say that the command line feels more simple than the Tortoise GUI. But who knows... I might change my mind.)

I can also see some "room" for other automatic versioning/backup solutions like AJC backup, autover or File Hamster (in addition to Mercurial). Committing changes is a voluntary action and, knowing myself, I'll obviously forget to commit important changes from time to time.  :) Or am I missing something ?

That's it for now... Thanks for helping me with my choice. Seems like Mercurial was what I was looking  for.
393
Announce Your Software/Service/Product / Re: Bvckup
« Last post by Armando on February 28, 2011, 08:07 PM »
Yes, it does. Modified the post to reflect that.

 I thought so... but wanted to make sure... 8)
394
Announce Your Software/Service/Product / Re: Bvckup
« Last post by Armando on February 28, 2011, 07:38 PM »
Great feature list. And... great Blog ! (I told you that already in your forum, but it deserves to be said again. :))

The Pipelined IO is intriguing. I can't wait to test V2 -- there seems to be lots of imagination, intelligence, care and effort in there!

BTW : does the File and directory move detection also mean... File and directory renaming detection?
395
Living Room / Re: Gmail accidentally resets some accounts
« Last post by Armando on February 28, 2011, 11:25 AM »
Actually, what are you doing with your backups?  Most of us take risks/have points of failure that we don't even realize.  And before you say that it's unlikely, I've had the perfect storm before.  RAID1 - both drives failed.  The latest backups were corrupted, and I didn't know (code red on my server).  I had to go back so far in my backups, that they were barely useful.

And are your servers truly more reliable than google's, when you look at it?  Or is it a false sense of security?  Not saying that a backup is not better than none... just that it's good to assess things after a while- that's the way that my perfect storm snuck up on me.

First, you have more experience than I do, wraith808. That's for sure. My point is just that it's better to have (Multiple) backups, even if Google's servers are theoretically much more reliable than any custom solution. (I can't empirically verify Google's safety for sure... However, I can verify (slightly "anecdoticaly")  that since 1988, none of my multiple backups failed all at once. )

Also, apart from pure hardware and software reliability, there's the people and the decisions they make. If I value my data, I'm more likely to value it more than Google is ever going to.

Then, when I travel, I don't always have internet connection so depending on Google's service alone is completely out of the question.
396
I wasn't speaking for ie in particular. We all know ie isn't the greatest. AFAIC though, I'm not sure exactly how ie was used in that specific case and if it does affect Outlook's performance. 2003 has the same problem depicted in yohimba's post, yet it's really speedy, no lag, nothing. I have 1.5b of stuff in there and I have yet to find something to manage e-mail, calendar, tasks, contact, etc. that's as responsive and isn't a performance hog (at all). :)
397
I hate it because IE is usually slow and they are using it as a shortcut for functionality, rather than programming their own interface. I shouldn't criticize too much because I don't know how much work is involved one way or the other.

...Especially if 99 % of users wouldn't notice the performance (or other) improvement but certainly wouldn't fail to notice the absence of the some other features they'd expect.  The 20/80 rule rules here. :)

And, yes, recoding everything all the time would most of the time be a lot more work (much more work). Components are there so that programmers don't always have to reinvent the wheel and spend their time coding in assembly, C and C++. Redesigning the wheel is sometimes appropriate, often it's not. I don't know how others feel about that...

Is the bug you noticed due to IE ? maybe... MrCrispy's explanation seems reasonable.
398
Living Room / Re: Gmail accidentally resets some accounts
« Last post by Armando on February 27, 2011, 11:20 PM »
"What else are you supposed to expect."

... Isn't this the perfect warning : don't trust "the cloud". Well, not too much. Backup your data yourself, have duplicates on site and elsewhere.

The best cheap and easy e-mail solution I found (a long time ago) was to have my own domain and forward everything to various e-mail services... and then download stuff to Outlook. So I'm not tied to any specific company and I can own all my correspondence. The only risk I'm taking is to loose my domain... Very unlikely.

I'm not familiar with the google apps for domains contracts, so I can't comment...
399
--managing multiple email accounts in just the right way
--locally stored files.  easily shared between applications.  Even Outlook is not as easy to share as people seem to think, but I don't want to get into those details
--quick templates; they let me automate a lot of things.
--layout control.  most other programs don't allow you to modify the layout as much as the Bat.
--powerful filtering.  gmail doesn't have it, nor does outlook.  Gmail's is way too simple.  Outlook is just odd and stupid.

Of course, I understand your obsessions. I'm pretty obsessive myself. However, I find that all the above can be done in Outlook well enough. Not to mention that it's much easier to find devices and services syncing with Outlook than with any other software. Yes, Gmail is way too simple for my needs also. That's why I don't manage my e-mail with it... I use Outlook to do so.

But... We obviously have different needs. And so I'd be curious to know what important action -- very specifically -- can be done in the bat which can't be done with Outlook...?

(Darwin : Hi ! :) )
400
I understand too the need to have all data locally. I do exactly the same.

But what do you need to do with your email, apart from classifying, sorting, filtering, searching and finding info in it?

Classify, sort, filter, search and find info in it - all the way back to 1995. I don't use Outlook, and if I only want to search my email, I find Thunderbird's searching ideal. I would not be happy having all of my email archive anywhere but on my local drive. It's over 4gb.





Hi timns. My last post was actually speaking to superboyac... :)
What I meant was : "what more than these functions do you need ?" (i.e. : Is the bat really necessary ?)
Pages: prev1 ... 11 12 13 14 15 [16] 17 18 19 20 21 ... 109next