Author Topic: Intelligent grouping and selection  (Read 5094 times)

Offline iEscape

  • Contributor
  • ***
  • Posts: 12
Intelligent grouping and selection
« on: June 11, 2008, 12:26:17 am »
Hey alt.binz users and dev ~

I'll start off this request by describing the actions I perform which led to my desire for these requests.  Every once in a while, when I'm starved for entertainment, I'll go to newzleech.com and search for something to watch.  I stock up on media in binges -- one huge download session and I'm set for 4-6 weeks.

One of my most common methods for finding things to watch is searching by group name.  I'll enter something like "vmt" in the 'Usenet Search:' field.  Then I set the advanced parameter, 'Size (MB):', to 2000 min and 5000 max.  This will return DVD-sized releases from VoMiT.  Next, I'll look through the nfos, check anything I want to watch, and 'open' the .nzb with alt.binz for downloading.  VoMiT is pretty prolific, so there might be 5-6 releases to download after a month.  That's about 25 GBs in DVD releases alone.  I'll look through other groups and their Xvid releases too.  My point so far is that I'm downloading collections into alt.binz with huge sizes -- sometimes up to 80 GB!

Very rarely will an 80 GB collection (labeled 'Subject' under the alt.binz columns) consist of one release, but quite a few different releases.  Each individual release will consist of 50-100 smaller files.  Thus, an 80 GB collection can contain hundreds and hundreds of files grouped together under the 'Subject' name.  This can be annoying to scroll through.

HERE'S WHERE MY REQUEST IS:It would be great if alt.binz was able to intelligently group releases in a further level of nested hierarchy under the 'Subject'.  For example, this is how alt.binz currently groups collections (I'll cheat and put a "..." instead of listing all the pieces.  I'll also *** the release names so I don't break any forum rules.):

Code: [Select]
Subject ▼
--------------------------------------------------------
 [-]-vmt
    ├ www.binaries4ever.net post [01/78] - "vmt-****-dvdr.nfo" yEnc
    ├ www.binaries4ever.net post [02/78] - "vmt-****-dvdr.r00" yEnc
    ├ www.binaries4ever.net post [03/78] - "vmt-****-dvdr.r01" yEnc
   ...
    ├ www.binaries4ever.net post [76/78] - "vmt-****-dvdr.r74" yEnc
    ├ www.binaries4ever.net post [77/78] - "vmt-****-dvdr.rar" yEnc
    ├ www.binaries4ever.net post [78/78] - "vmt-****-dvdr.sfv" yEnc
    ├ Most Wanted presents : vmt-*****-dvdr Binaries4Ever.net [01/92] - "vmt-*****-dvdr.nfo" yEnc
    ├ Most Wanted presents : vmt-*****-dvdr Binaries4Ever.net [02/92] - "vmt-*****-dvdr.r00" yEnc
    ├ Most Wanted presents : vmt-*****-dvdr Binaries4Ever.net [03/92] - "vmt-*****-dvdr.r01" yEnc
   ...
    ├ Most Wanted presents : vmt-*****-dvdr Binaries4Ever.net [90/92] - "vmt-*****-dvdr.r88" yEnc
    ├ Most Wanted presents : vmt-*****-dvdr Binaries4Ever.net [91/92] - "vmt-*****-dvdr.rar" yEnc
    ├ Most Wanted presents : vmt-*****-dvdr Binaries4Ever.net [92/92] - "vmt-*****-dvdr.sfv" yEnc
    ├ #alt.binaries.dvdr: The.*******.2006.NTSC.DVDR-VoMiT "vmt-*******-dvdr.nfo"
    ├ #alt.binaries.dvdr: The.*******.2006.NTSC.DVDR-VoMiT "vmt-*******-dvdr.par2"
    ├ #alt.binaries.dvdr: The.*******.2006.NTSC.DVDR-VoMiT "vmt-*******-dvdr.r00"
   ...
    ├ #alt.binaries.dvdr: The.*******.2006.NTSC.DVDR-VoMiT "vmt-*******-dvdr.r93"
    ├ #alt.binaries.dvdr: The.*******.2006.NTSC.DVDR-VoMiT "vmt-*******-dvdr.rar"
    ├ #alt.binaries.dvdr: The.*******.2006.NTSC.DVDR-VoMiT "vmt-*******-dvdr.sfv"
    ├ www.binaries4ever.net post [1/9] - "vmt-****-dvdr.par2" yEnc
    ├ www.binaries4ever.net post [2/9] - "vmt-****-dvdr.vol000+01.PAR2" yEnc
    ├ www.binaries4ever.net post [3/9] - "vmt-****-dvdr.vol001+02.PAR2" yEnc
   ...
    └ (rest of the pars)

And this is how I'd prefer to see it grouped:
Code: [Select]
Subject ▼
--------------------------------------------------------
 [-]-vmt
    ├ [+] www.binaries4ever.net post - "vmt-****-dvdr" yEnc
    ├ [+] Most Wanted presents : vmt-****-dvdr Binaries4Ever.net - "vmt-*****-dvdr" yEnc
    ├ [+] #alt.binaries.dvdr: The.*******.2006.NTSC.DVDR-VoMiT "vmt-*******-dvdr"
    └ [+] PARs

As you can see, each release is in its own expand/collapse collection.  The names of the collections are somewhat arbitrary.  I simply removed the parts of the names where the individual file names had differences -- usually the file number brackets (e.g., [76/78]) and the file extensions (e.g., .nfo, .rar., .sfv, .r00, .r01, etc.).  If possible, you could further simplify the collection names by just using the file within the quotation marks and removing the file extension, which would look like this:

Code: [Select]
Subject ▼
--------------------------------------------------------
 [-]-vmt
    ├ [+] vmt-****-dvdr
    ├ [+] vmt-*****-dvdr
    ├ [+] vmt-*******-dvdr
    └ [+] PARs

This is very clean looking, but I'm not sure if this naming convention is common to all NZB sites and/or releases.  I suppose the easiest way to group them would be to ignore renaming the nested collection altogether, and just give it the name of the topmost file in the release.  It would be very nice if you added an option to rename the nested collections, though.


NOW FOR THE OTHER REQUESTS:It would be awesome if the nested collections were draggable under the top collection (Subject), so you could quickly reorder the download order.  It would also be great if you could perform the same actions on nested collections that you can currently perform on collections.  Such as: delete them, split and group them, select them with SHIFT and CTRL for ranged and individual selecting, pause and unpause, etc.


I realize that this might be difficult to implement.  If it is very difficult, or you don't like the idea, I hope you might consider adding something for me as an alternate or stopgap measure.  I'd like to see a 'Select all' option in the context menu.  Sometimes, I have to go through the hundreds and hundreds of files in a collection and pause a large range of them, because I'm running out of hard drive space.  After pausing parts of a collection, I'll go to sleep and let my hard drive fill up.

When I wake up and unRAR them to my secondary drive, I delete the RAR/PAR sets and make room to unpause and download the rest of the collection.  However, the collection itself is not paused, the files inside it are.  The only way to get them started again is to select the range of them and unpause the files within the collection.  You can do this by clicking the topmost file, then scrolling down to the bottommost file, holding SHIFT and clicking on the bottom file to select the whole range, then right-clicking in the range and choosing unpause.

I'd prefer that a 'Select all' option existed instead.  Right-clicking on a file within a collection and choosing 'Select all' would select all the pieces in the collection.  You could then right-click on the selection and unpause it, skipping the keyboard input altogether.  I realize I could split the group and pause the split pieces, but the 'Select all' method is more intuitive.  If you don't think it's more intuitive, you might at least concede that it's a more common, widespread method.

Anyway, I'll stop rambling.  Sorry for being so verbose.  And thanks for reading and considering my request!

Offline Hecks

  • Contributor
  • ***
  • Posts: 2011
  • naughty cop
Re: Intelligent grouping and selection
« Reply #1 on: June 11, 2008, 12:53:34 am »
Heh, verbose is fine with me, it's always good to see someone thinking long and hard about their requests :)

I'm sure there'll be discussion about this one, so let me throw in my 2c worth: it strikes me that the nub of the issue here is the way that you add items to the queue.  Alt.Binz works best when you don't just dump stuff into one NZB node (aka 'collection'), but instead make each node a distinct post.  There are numerous advantages:

1. ability to sort and rename stuff properly
2. PAR2 reparing/unrar only triggers when collection is complete
3. ability to pause individual downloads
4. ability to move downloads up/down the queue
5. ability to set dl & unrar directories per dl
6. ability to set different execute on completion commands
etc. etc.

So basically, in the nice example you give, I'd have taken just a few more minutes to add that stuff separately to the queue.  It saves a LOT of hassle in the long run.  This is also how I think 99% of Alt.Binz users actually do it, but maybe I'm wrong.

As for searching in the way you describe, maybe there are better / quicker ways to find stuff you're interested in.  For instance:

http://www.vcdquality.com/index.php?genre=5
http://www.nzbs.org/index.php?action=browse&type=2

Anyways, that's enough from me. :)

P.S. You really watch VoMiT releases? lol
« Last Edit: June 11, 2008, 12:56:38 am by Hecks »

Offline iEscape

  • Contributor
  • ***
  • Posts: 12
Re: Intelligent grouping and selection
« Reply #2 on: June 11, 2008, 02:02:51 am »
Hey Hecks, thanks for the feedback.  I read your points and I see what you're saying, but I think if the auto-subcollection idea I proposed was implemented properly, it would essentially be the same as downloading releases individually.  Thus, all the points you made would/could apply to a collection of subcollections if coded properly.

Since my main search avenue is newzleech.com, and I have a set of decent search terms laid out, I find my method pretty easy.  It's really newzleech that makes it so easy the way they organize things.  I search my term with any advanced parameters, check off everything I want off the list within the last 30 or so days, then start downloading.  I think I save a lot more than a few minutes that way.  If I were to switch back and forth between Firefox and alt.binz, checking individual files, starting the download, renaming things, unchecking things, etc., I'm pretty sure it would add hours to the process, considering the sheer number of things I download at one time.

I used to download releases as they came out, but I found I was putting far to much time and effort into it.  I'd see something pop up on vcdquality and search for it immediately on newzleech.  The two sites don't always synch up, time wise.  That got annoying (not to mention the naming conventions of some groups making some releases a challenge to find) eventually and I switched over to the wait period between download sessions.  Now I just spend a few hours every month searching for downloads I want and a day or two downloading and unpacking them.  It's a lot more laid back and I don't have to devote my life to "the hunt."

I don't use the RAR/PAR features of alt.binz as I've never gotten them to work how I'd like.  I know some improvements have been made in the last few revisions, so I'll check into that again.  It would be nice to be able to skip manually PAR checking and unRARing everything.

I have ten DVD+RWs which I fill with content, watch, delete the content from my drive, then refill the DVD+RWs until I'm out of content.  I'm not really concerned with archiving anything, because I eventually buy what I want to own.  I'll be damned if I give the rental places any more of my money though.   ;)

I do watch VoMiT releases.  I'll watch any group's release.  By the time I get around to my monthly hunt, the propers are usually out anyway.  The reason I used VoMiT in my initial post is that they're the easiest group to search on newzleech, pretty much -- much easier than a group like DREAMLiGHT.  Try entering "dl" and see how many bad hits you get compared to "vmt".   ;D   (I wish newzleech had more stringent matching, so "dl-" could be matched.)

Lastly, thanks for the link to nzbs.org.  I didn't know about that site and it looks to be free and gives out the short release name, unlike vcdquality.  I'd been researching my release names on newzbin.com until they recently cut that off to non-members.  I switched to nzbmatrix.com for getting the short name, but the "Popular Searches" box on that site makes my eyes bleed.

Offline Hecks

  • Contributor
  • ***
  • Posts: 2011
  • naughty cop
Re: Intelligent grouping and selection
« Reply #3 on: June 11, 2008, 02:37:26 am »
I'm not surprised PAR/unrar isn't working well for you.  If you have a single node in the queue containing 80GB of stuff, Alt.Binz will first of all dl the small PAR2s - all of them - then wait until the whole collection is empty before trying to repair or unrar anything ... and then all at the same time.  This way, you also lose the convenience of managing dl/decode/par2 threads, naming destination folders after collections/NZBs, and so on.

Everyone has their preferred method of searching, I suppose.  Personally these days, I just stick with sources that use req numbers or post IDs or similar stuff that's easy to search for ... then it's just select > right-click > add to queue, changing folder info if needed and/or using Alt.Binz's automatic release name matching.  Otherwise it's autodownloading with RSS, which is a great way to add stuff to your queue (as Paused is optional) for later browsing - all as separate queue items.

Newzleech also has a Posts section, where users merge stuff into collections, can't say I've ever used it much myself, though.  Nzbs.org is all automated, so expect a delay of 2/3 hours from posting to stuff appearing there.  Also, you need to configure the RSS after you've registered if you want to use that for direct downloading.  Newzleech and Binsearch and just about every other decent indexing site have RSS feeds, and you can use a feed aggregator/pipe to merge them.



« Last Edit: June 11, 2008, 02:38:57 am by Hecks »

Offline EyeBaller

  • Contributor
  • ***
  • Posts: 66
Re: Intelligent grouping and selection
« Reply #4 on: June 11, 2008, 05:09:31 am »
If I were to switch back and forth between Firefox and alt.binz, checking individual files, starting the download, renaming things, unchecking things, etc., I'm pretty sure it would add hours to the process, considering the sheer number of things I download at one time.

Just wanted to clarify why you're switching to firefox? Are you searching newzleech from FF, saving NZBs and importing them? Why not use the built in newzleech tab in Alt.Binz. Then your process becomes:
1. Search with your parameters (in Alt.Binz newzleech tab)
2. Scroll to release, right click to view nfo (again, right in Alt.Binz)
3. Yes I want to download this, right click the collection again and add to queue
4. Scroll down and repeat

It's much faster than going through the website.

(PS: Although you can also select multiple releases in the Alt.Binz newzleech tab and you will get the same result (everything added as one "nzb") so your request is still a good one.. kinda...)

Offline iEscape

  • Contributor
  • ***
  • Posts: 12
Re: Intelligent grouping and selection
« Reply #5 on: June 11, 2008, 08:48:41 am »
Just wanted to clarify why you're switching to firefox? Are you searching newzleech from FF, saving NZBs and importing them?
Yep, I'd gotten so used to using the web browser for searching, I haven't been using alt.binz to its full potential.  It's a shame too -- my first request on the forum was to ask rdl to fix the proxy options for search engines so I could use my socks5 proxy connection for the newzleech tab.  He fixed that for me within a week, but I slid back into my Firefox habit.   :-[

I've just been reexploring the search tabs and I can see how much easier it is.  I don't want to withdraw my request just yet though.  I still like selecting a range of downloads in one sweeping move.  After some experimenting, I discovered that newzleech.com will only allow you to form a collection of 2000 or less files in one NZB.  This limits the number of releases that can be selected, although alt.binz doesn't tell you this when you attempt to assemble a 2001+ file collection.  It simply does nothing.


Thus, I'll propose some ideas where the newzleech tab might be improved.  At the very least, there should be a warning when trying to assemble a 2001+ file nzb so we know nothing is going to happen after trying to add it to the queue.  :P

Since the newzleech tab has a column dedicated to the number of files per release, it would be handy if there was an info bar or status bar that showed how many releases and how many overall files you had selected.  Any other pertinent information could be displayed in the bar, though I'm drawing a blank for ideas as to what.  Oh, just thought of this -- the current size of your total selection and the free space of your drives next to it.

I also noticed on downloading a NZB with so many files, it takes a while.  It would be nice if there was a progress meter for NZB downloads.  I guess a collection as large as 2000 files is close to a 20 MB NZB file.  I had to pause the releases I was downloading to get alt.binz to start downloading this large of a NZB.  Does alt.binz currently give download priority and reasonable bandwidth to new NZB downloads (i.e., adding new files to the queue)?

Aside from the information bar, I had a few more ideas.  If you go to the newzleech tab and enter "replica" as a search term, you'll currently get two pages of hits.  Why bother with pages?  I'd love the option to download all the pages at once and peruse one long page.  This way you could make a range selection across multiple pages, which would be awesome.  I suppose some search terms could match a lot more than two pages, so perhaps an option to limit the number of pages grabbed?  I do hope grabbing multiple pages wouldn't have too negative an impact on the newzleech server.

The next idea is one to bypass newzleech's 2000 file limit in one NZB by having alt.binz treat a large selection of releases as a collection of individual releases internally, thus downloading the NZB for each release separately.  I don't know how this would effect newzleech either -- downloading eighteen 1 MB NZBs as opposed to one 18 MB NZB file, I mean.  I also don't know if this might be a way to better avoid corrupted NZB downloads.  I can't think of the last time I downloaded a corrupt NZB anyway.

Of course, using the method of large selections of files, you might want a more persistent way to select files.  Something like the check boxes on the newzleech web site.  Or perhaps a new pane in the tab so when you double click a release, it would transfer to the 'selection pane' and be removed from the 'search pane'.  This way you could examine the individual releases for their NFO and IMDb links without altering the selected range of files.  You could also add a bar of buttons at the top of the 'selection pane' such as, 'add to queue', 'add as first', etc.

I still really like the idea of subcollections under a main collection.  It just further allows you to organize your downloads and makes it easier to keep a clutter-free display of your downloads possible.  So, to sum the concept up, using the previous ideas:

I'd go to the newzleech tab, enter "replica" in the search field and click 'Search' (or hit Enter).  I'd get a full list back of all hits, without any page separation.  I'd click through the NFOs and check off the releases I wanted to download.  I'd right-click to choose 'add to queue' or 'add to queue (paused)'.  A box would pop up similar to to current with the following differences.  The 'Enter Name' field would be of the search term I entered, in this case "replica" -- this could be renamed, of course -- and this is the name of the main collection.  There would be a choice on this page to either cycle through all the individual releases and choose to rename them (these would become the subcollection names under the main collection) and choose their download and unRAR directories individually, or a choice to let alt.binz perform that step automatically.


Anyway ... crazy, late-night brainstorming session over.  I hope I made some sense and these ideas aren't too useless.  I like the subcollection idea for organization and I like the large range selection of releases for streamlining the process with (hopefully) less clicking and less overall effort involved.

On that note, I'd really appreciate the light you guys have shed on some features I've overlooked and been misusing in alt.binz.  I'd further appreciate any more suggestions to streamline my searching process.  I came up with the search by release group name as my best option and settled on it, but it sounds like you have a pretty streamlined methods, Hecks.  If you'd be willing to share, I have a feeling you could help simplify my life a bit more.  Every little bit helps!  ;)

Offline Hecks

  • Contributor
  • ***
  • Posts: 2011
  • naughty cop
Re: Intelligent grouping and selection
« Reply #6 on: June 11, 2008, 09:51:52 am »
1. Status bar already shows total size of queue and free space on dl drive.

2. IIRC, Alt.Binz does download an NZB if Newzleech > 2000 posts, it just contains the html error message. I suppose that could be error-checked.  :)

3. Pages: Alt.Binz just follows the site in this respect.  To get more results per page you need to be logged in to the site, unfortunately.  Sending multiple requests to the site to dl every page in the possible range is a really bad idea, I think.  Just narrow your search by using better search terms. ;)

4. The rest of your suggestions - progress indicator for NZB downloads, etc., are again all based on the assumption that the  majority of users will dump large selections into one NZB instead of adding individual posts separately.  I'm afraid you might be in the tiny minority here with that ... and since the time/effort required for you to learn to use Alt.Binz in the 'proper' way is a lot smaller than the time/effort probably required to implement your requests, I somehow doubt that they'll be implemented anytime soon.

Still, interesting read, thanks. :)


Offline awGole

  • Contributor
  • ***
  • Posts: 18
Re: Intelligent grouping and selection
« Reply #7 on: June 11, 2008, 01:16:36 pm »
I made a similar request a long time ago now.. so +1 for sub-collections (or a user controlled "top-level" or whatever i called it hehe).

Offline EyeBaller

  • Contributor
  • ***
  • Posts: 66
Re: Intelligent grouping and selection
« Reply #8 on: June 11, 2008, 04:49:08 pm »
Thinking about it some more the "intelligent grouping" definitely would be a good idea. I most commonly run into this problem when I want to download multiple episodes of tv shows and add them all as one nzb, so +1 for that.

(again.. just really my laziness of not adding each episode individually)

Offline iEscape

  • Contributor
  • ***
  • Posts: 12
Re: Intelligent grouping and selection
« Reply #9 on: June 11, 2008, 08:17:06 pm »
1. Status bar already shows total size of queue and free space on dl drive.

Yep, the main status bar does, but I was thinking of an information bar that coincides with the multiple file selection you might be working on creating.  This would be more of a convenience feature, having pertinent info (the size of the current selection right next to your available free space along with other info) grouped together, all based on the task you're currently performing.  Also, it would be nice if all of your drives free space were shown, so you could more easily decide where to download or unRAR things.

2. IIRC, Alt.Binz does download an NZB if Newzleech > 2000 posts, it just contains the html error message. I suppose that could be error-checked.  :)

I tested this and it doesn't seem to work.  Using the example I used in my last post, do a search for "replica".  Sort the hits by the files column.  Now select 20 of the files with >100 files in them and 'add to queue (paused)' -- nothing will happen.  Now deselect a release or two using CTRL + a mouse click.  Once the total file count goes below 2001, the NZB will be created (it will just take a while) after choosing 'add to queue (paused)' again.  I thought this was alt.binz's fault at first, but I went to the newzleech site, used the same search and selections and it threw up an error, unfamiliar to me, explaining the 2000 file cut-off.

I think the best way to get around this is by treating a multi-selection as many individual releases during the NZB download phase.  I did a mini-test, downloading two NZBs separately and then combining them into one NZB, and it seems that they're pretty close to equal in overall download size.  The combined NZB is more efficient only by a few hundred bytes in my test.  If this combined file size efficiency is greater than I realize, perhaps a more intelligent separation of multi-file NZB could be worked out.

3. Pages: Alt.Binz just follows the site in this respect.  To get more results per page you need to be logged in to the site, unfortunately.  Sending multiple requests to the site to dl every page in the possible range is a really bad idea, I think.  Just narrow your search by using better search terms. ;)

True enough.  I was worried about this one as I don't want to bombard the last remaining free databases with spider-like activity.  However, maybe a reasonable waiting period between page fetches to build a mutli-page display could be worked out?  The reason I ask this is because newzleech is not exactly built for search accuracy (unless you're search by post REQ number as you do Hecks).

Like I mentioned earlier, if I wanted to search for a group like DREAMLiGHT's releases, I'd enter "dl" into the search field.  Further restrict the search by entering a min file size of 3000 and a max of 5000 (since DREAMLiGHT puts out DVD-sized releases), and you'll get 8 pages of hits back currently.  Most of these matched files are not released by DREAMLiGHT.  Unfortunately, newzleech won't match "dl-" to make this search more productive -- it ignores the hyphen and most other non-letter characters.  On the other hand, alt.binz will allow you to more precisely match the hits you receive against a second term, referred to as the 'Filter subject' field.

If I entered "dl-" into the 'Filter subject' field, I'd get a set of matches closer to what I wanted.  I still get some bad matches, so adding a quotation mark to the front of that "dl-" ("dl-) will get me exactly what I want.  On the negative side, I only see eight good matches now, all on the first page.  There are seven more pages that might contain matches, but I'll have to flip through them.  This makes a multi-selection impossible across pages.

The two ways I can think of to fix this are:

  1.) Downloading all the pages (in a reasonable way) for a single display through alt.binz.  Also, I see this as a possible benefit for newzleech as you won't be clicking back and forth between pages.  This could save newzleech some bandwidth, unless alt.binz actually caches these search pages instead of refetching them.

  2.) Implementing the 'selection pane' I proposed in my last post.  This would let you click through the pages, look through the NFOs and IMDb info at your leisure, add the releases you want to the 'selection pane', then add your selection to the queue under one collection when you're finished.  I know you could add these files to the queue singly, but things could get cluttered quickly that way.  A 'selection pane' has the added benefit of being able to hold selections in place while performing different searches.

Think of how useful this would be for a task like making a mix of music to listen to.  You could search for a bunch of different songs, move them over to the 'selection pane', then download them in one collection.  You could do this with individual songs, and organize them into the same download folder, but I think that would be a lot more work.

4. The rest of your suggestions - progress indicator for NZB downloads, etc., are again all based on the assumption that the  majority of users will dump large selections into one NZB instead of adding individual posts separately.  I'm afraid you might be in the tiny minority here with that ... and since the time/effort required for you to learn to use Alt.Binz in the 'proper' way is a lot smaller than the time/effort probably required to implement your requests, I somehow doubt that they'll be implemented anytime soon.

Still, interesting read, thanks. :)

I think a progress indicator for NZB downloads would not only be beneficial to a multi-file selector, but also to the single-file user.  A single release can easily have a NZB well over 1 MB, so I'm guessing can take a decent amount of time to download for someone with slower-than-optimum-broadband.  For example, a recent HD release of the movie Superbad has a NZB size of 3.27 MB.  Now, factor in that your currently running downloads are using all or most of your bandwidth.  Adding files to the queue could seem to take forever.  Without a progress bar of some kind, you might doubt that they're being added at all (until they finally pop up). 

Lol, when I was trying to create the 2001+ file NZBs from an alt.binz newzleech search, I thought my NZB was possibly hundreds of MBs and I just had to patiently wait it out until the collection popped up.  Then I thought the current download was stealing all the bandwidth so I paused the downloads.  Then after another two minutes, the jig was up!

I don't think it's that uncommon for people to create collections with multiple releases.  I've done it a lot in the past -- especially for TV seasons that I missed (grabbing most of, or an entire season in one go), or an entire series of TV shows (grabbing several seasons of a series on DVD in one go).  I think downloading music by artist applies here too.

I suppose it boils down to how you prefer to search for things.  If you search specifically for one thing, by post number or by specific name, a lot of my suggestions wouldn't help you out.  However, if you search by release group name, or by artist name, or by series name and grab a lot of stuff at once, I think there is room for improvement in alt.binz here.  Hopefully, my suggestions might help out.  Thanks for the feedback and challenging responses everyone.   :)


P.S. - One thing that I forgot to mention when describing the process in my last post.  On the initial dialog that pops up for your main collection, you'll most definitely want to have the option to choose to sort the releases in your collection either directly under the main collection, or create subcollections based on the release names under the main collection.  There are certain times when I'd rather put a multi-file collection in one folder and see it visually displayed in one list in the download queue, and other times when I'd like to separate a multi-file collection into different folders and see it displayed in the download queue in a more hierarchical structure for easier manipulation.

P.P.S. - Are my posts getting longer and longer?   :-[

Offline Hecks

  • Contributor
  • ***
  • Posts: 2011
  • naughty cop
Re: Intelligent grouping and selection
« Reply #10 on: June 12, 2008, 03:16:04 am »
As I said, Alt.Binz will create an NZB if you select > 2000 from Newzleech, and will save it to your NZB folder.  But it will only contain the error message, and so can't be imported to the queue.  That is something that could be error-checked.

But I've been repeating myself quite often in this thread so far, so that's it from me for now.

P.S. Rdl is the only developer of Alt.Binz, in case this was unclear.

P.P.S.  And in case this is also unclear, the way that you currently search for results is quite possibly the worst way of doing it that I can imagine.  I think the likelihood of anything being implemented to compensate for your inability to search properly is almost zero.

That said, an Autosplit function might be useful to others in different scenarios, if used along with Join.  Quickly splitting seasons, for example. :)
« Last Edit: June 12, 2008, 03:24:13 am by Hecks »

Offline iEscape

  • Contributor
  • ***
  • Posts: 12
Re: Intelligent grouping and selection
« Reply #11 on: June 12, 2008, 04:10:16 am »
As I said, Alt.Binz will create an NZB if you select > 2000 from Newzleech, and will save it to your NZB folder.  But it will only contain the error message, and so can't be imported to the queue.  That is something that could be error-checked.

But I've been repeating myself quite often in this thread so far, so that's it from me for now.

Yeah, I read and understood what you wrote.  Our definition of a NZB might differ though.  Just because you slap on a .nzb extension to a file doesn't make it a NZB.  Thus, a HTML error message written into a text file with a .nzb extension tucked away into the NZB folder doesn't mean that "Alt.Binz does download an NZB if Newzleech > 2000 posts" ... to me, anyway.  But, yeah, an error definitely needs to be thrown up for the alt.binz user to see in that situation.  Unless, of course, a more graceful solution was developed, like alt.binz handling the error transparently and splitting the NZB for you.

Oh, and sorry that you've had to repeat yourself so much in this thread on my account.

P.S. Rdl is the only developer of Alt.Binz, in case this was unclear.

Nope, not unclear to me.  I did write "alt.binz users and dev" in my first post, and mentioned in another post how Rdl fixed something for me in a later post.  If you think I'm going on and on because I think you can personally get something done, that's not the case.  I just thought this was the place to have an open discussion about requests.  I was truly just brainstorming, and trying to improve and embellish upon the idea.

P.P.S.  And in case this is also unclear, the way that you currently search for results is quite possibly the worst way of doing it that I can imagine.  I think the likelihood of anything being implemented to compensate for your inability to search properly is almost zero.

LOL!  Thanks for the heady compliment.  :-X  My method has worked for me pretty well so far.  I much prefer it to trolling vcdquality.com and grabbing things as they're released.  It's also a crapton better than waiting a month and scrolling through a month-long worth of posts on vcdquality, looking each up and downloading the NZBs individually.  I know I could improve my method, though.

If you read my posts, I did ask for some guidance in searching.  I guess you might not be willing to share?  You could give generic advice like, 'use RSS' or 'use REQ numbers', but that's not really too helpful.  I've seen REQ and post numbers all over the place, and I know what RSS is -- I just don't know how to use them efficiently.  I've done a bit of research in trying to streamline my process, but I'm just not savvy enough to find the resources, and I'm not willing to devote (what I consider) an insane amount of time trying to figure it all out on my own.  The last thing I'll be doing is downloading headers ... HA ... I really appreciate the legwork people do to index releases and put the NZBs together.  I show my appreciation by using the NZBs and alt.binz.


You almost seem hostile to the idea of large collections, Hecks.  Why is that?  I figure that sites like newzleech.com and programmers like Rdl added the ability to select more than one file at a time for a reason.  Just because I'm using the ability to download DVDs instead of songs doesn't make it bad way to hunt and gather.  That's my opinion anyway.

Peace!   ;D
« Last Edit: June 12, 2008, 04:12:32 am by iEscape »