• Please use real names.

    Greetings to all who have registered to OPF and those guests taking a look around. Please use real names. Registrations with fictitious names will not be processed. REAL NAMES ONLY will be processed

    Firstname Lastname

    Register

    We are a courteous and supportive community. No need to hide behind an alia. If you have a genuine need for privacy/secrecy then let me know!
  • Welcome to the new site. Here's a thread about the update where you can post your feedback, ask questions or spot those nasty bugs!

New Posts suggestion for improvement

Asher Kelman said:
I am constantly looking at adjusting out threads and fora. Whenever you see something out of whack, let me know.

Asher
Not exactly "out of whack," but I don't use the "New Posts" or "Quick Links->Today's Posts" (which I believe are the same page) because I strongly prefer the list of threads containing new posts be ordered by Forum name rather than by date. That is, I want to see all the new posts in a Forum before I mentally shift perspective to another forum's subject area.

The Advanced Search page supports both limiting a search to articles posted since my last visit and ordering the search result by Forum, but requires something be entered in either the "Search by Key Word" or "Search by User Name" field. Searching for "a" or "the" seems a bit absurd, not to mention expensive (I suspect) in terms of server resources.

Have I missed something that changes the order of the "New Posts" or "Quick Links->Today's Posts" result? If not, please consider adding an option to the "New Posts" or "Quick Links->Today's Posts" result page that enables users to choose the order of the result. Since the Advanced Search already offers the option, the change would seem to require connecting the user interface to existing code, i.e., a relatively modest change.

Bob
 

Asher Kelman

OPF Owner/Editor-in-Chief
Bob,

I'm currently thinking of paradigms for easier navigation. If we know what is wanted, then I may put a bunch of these ideas together to do it as one item.

This feedback is helpful. One thing one does want avoid doing is lots of changes to lots of things in lots of different ways. That might make upgrades more expensive to upgrade to. I'd like to collect a bunch of ideas then try to build them together if that is possible or find a commercial addon that already has that capacity. My current focus is getting a gallery going and to make it easier for Tim Armes to run the retouch forum by hosting the images.

Asher
 
Rob Peterson said:
Not exactly "out of whack," but I don't use the "New Posts" or "Quick Links->Today's Posts" (which I believe are the same page)
They are not the same. One is based on cookies* and your last visit while one is temporal.

my $0.02 USD (my two cents**),

Sean

* It could also be based on a server side stat but the end result from a user perspective is the same.

** A USA based cliche on the value of a small comment.
 
Asher Kelman said:
ne thing one does want avoid doing is lots of changes to lots of things in lots of different ways. That might make upgrades more expensive to upgrade to.

Hi Asher,

<Sean being contentious and opinionated as usual ;) >

While having source code is nice for customization and I appreciate the effort going into it, I should also warn that extensive customization is expensive in terms of applying security patches. Even if you are not paying the admin, doing source code diffs is a time intensive (i.e., expensive) operation and errors can cause security issues.

sincerely,

Sean (speaking as a software engineer and rather critical as I have been tonight <smiling fondly>)
 

Asher Kelman

OPF Owner/Editor-in-Chief
Sean,

So you agree with me!

However, explain "doing source code diffs". Tell me what is "diffs"?

Asher
 
Asher Kelman said:
Sean,

So you agree with me!

However, explain "doing source code diffs". Tell me what is "diffs"?

Asher
I, too, write software for a living.

Someone using the term "diffs" usually is a software person with a bit of Unix in his background. The Unix diff command displays how two similar files differ by showing, typically, only added, removed, and changed lines. Some programs now display differences using modern graphical display, e.g., most integrated development environments.

The "diffs", in this case, are the differences between the code the author distributes and the code you're running. Said another way, your changes to the original code.

The concern is that applying local changes to revisions issued by the author in order to maintain your local enhancements becomes increasingly difficult as the code evolves. Every author release requires you to examine the diffs (differences); decide if the set of changes supporting each of your enhancements remians valid; apply the valid changes; test the entire modified product to ensure nothing broke and that the changes didn't introduce security risks; and if the product passes the tests, install the update. The amount of work increases rapidly as the number of local changes increases.

If the author releases patches between full source releases, frequently to fix a security vulnerability, you must check that the patch doesn't conflict with your local changes. Doing so requires executing the above process, including the tests, before moving the patched code into production.

In an ideal world you would provide the author your enhancments as both specifications and as implemented code, and the next release wouldn't require you to (re)apply any local changes because the new release incorproates your enhancments. And in that same ideal world, the author would never need to publish patches.

Bob
 
Asher Kelman said:
So you agree with me!

Hi Asher,


Yes. Sometimes even I am agreeable. <silly smile> The question at hand is whether or not too many changes to enahance usability will make the site too difficult to maintain and keep secure (apply vbulletin security patches without breaking the custom changes). And since the increases in complexity tend* to be super-linear if not exponential based on the number of changes too many changes can become very expensive. A good software architecture should reduce this increase in complexity, but it will never solve it all.

enjoy,

Sean



* I am ignoring superficial changes like colors, fonts, names, logos, and etcetera here.
 
Rob Peterson said:
Someone using the term "diffs" usually is a software person with a bit of Unix in his background. The Unix diff command displays how two similar files differ by showing, typically, only added, removed, and changed lines. Some programs now display differences using modern graphical display, e.g., most integrated development environments.

The "diffs", in this case, are the differences between the code the author distributes and the code you're running. Said another way, your changes to the original code.

I would exclude the Unix reference, but since I resemble it I may not be a neutral observer on this one (broke my teeth on Ultrix which dates me for those who know).

As to the general issue of diffs (differences), the only thing I would add to this explanation is that diffs and merges* are more of a facet of version control (retaining every version of a file created**) which is used to take control of dealing with both changes over time and parallel development by multiple developers.

enjoy,

Sean


* Comparing a custom file with a patch/update and preserving customizations while updating the overall software.

** Technically checked in.
 
Sean DeMerchant said:
I would exclude the Unix reference, but since I resemble it I may not be a neutral observer on this one (broke my teeth on Ultrix which dates me for those who know).

Sean
I first read Usenet newsgroups on an 11/750 running Ultrix. It was named Waltz because a guy named Walt successfully pushed for it, configured it, installed the software, and managed it. Yes, that's been a while (> 20 years).

While diff and merge most commonly are found supporting versions, diff finds other uses, such as comparing a directory with what should be a backup copy. The differences are the files present in one directory but not the other.

One can also think of a variant of diff showing differences in the edits applied to an image file. This should be possible in tools that record changes as a sequence of transformations instead of an updated file. This is as if you could record the entire history of Photoshop edits leading from a single source file to two different results, then looking at the differences in the command sequence. We're trying to do something akin to this in the Retouching & Repairing forum by documenting the steps leading to a result. (There! I got back on topic! :cool:= ).

Bob
 

Asher Kelman

OPF Owner/Editor-in-Chief
Thanks guys for guiding me.

On this and other websites, there a lot of locations to visit and for different or overlapping reasons. Also I am planning galleries. So I look for simple ways of implementing changes, but which don't cause scattered issues which we'll need to trouble shoot.

So far, we've well. We haven't as yet screwed up, except that sometimes the quick comment post hangs up and one has to use the Advanced option. However, I don't know that this is our fault!


Asher
 
Top