ticalc.org
Basics Archives Community Services Programming
Hardware Help About Search Your Account
   Home :: Archives :: News :: The Quality of Our Archives

The Quality of Our Archives
Posted by Michael on 4 March 2004, 22:29 GMT

You may have noticed the low numbers of new files added to our archives. We're having a debate about what to do with a growing problem: Programs that simply aren't very useful to anyone. There are more quadratic solvers in our archives than should ever exist, notwithstanding the fact that most models have this as a built-in feature. Our possible solutions are:

Currently, all files that meet the site policies are processed and uploaded to our archives. Since this doesn't seem to be working well, here are the ideas under consideration:

  • The file archivers could manually screen programs for those deemed "junk", in the sense that they lower the signal-to-noise ratio of our archives rather than increase it. Authors would have to e-mail an appeal for rejected programs. This would cause a longer waiting time for processing files.
  • We could implement a rating system and organize programs by rating. This allows all programs to remain on the site, but the most useless could be filtered out. A method of dealing with new programs and low/high numbers of votes would have to be developed.
  • Our current folder system stops at games, programs, math, et cetera. For ease of browsing, this could be expanded to sub-categories like games/board, games/shooter, and games/guessthenumber. This doesn't limit the number of files added, it only categorizes them so folders are more concise and relevant.
  • Lastly, we could just leave everything as it is now.

We're asking for your input on what to do. There is a survey posted in conjunction with this article where you can vote on this issue. Thank you.

Update (Archiver): We will not be deleting files (at least not this time around), that was never one of the options. If you do want some of your programs deleted e-mail filearchive@ticalc.org.

  Reply to this article


The comments below are written by ticalc.org visitors. Their views are not necessarily those of ticalc.org, and ticalc.org takes no responsibility for their content.


Re: The Quality of Our Archives
vadim p  Account Info

this is a bit off-topic, but what i really have is when people release a couple of versions of the same game, and unless the put the version #, it's really hard to figure out which one is the latest, so maybe it should be made a requirement?

anyways I agree with points #2 and #3, because they both would ease off the searh-for-a-good-game process :)

Reply to this comment    5 March 2004, 12:38 GMT


Re: Re: The Quality of Our Archives
Lewk  Account Info

Maybe its just me but i have no clue what Pheonix version to download.

Reply to this comment    6 March 2004, 04:34 GMT


Re: Re: Re: The Quality of Our Archives
PlaidPhantom Account Info

I downloaded all of them for my different moods.

Reply to this comment    6 March 2004, 06:08 GMT

Re: The Quality of Our Archives
Anders Tiberg  Account Info

I've noticed that bad programs often get more downloads than good ones because their description is very long and very positive. I don't think you can solve the problem with quality any other way than to have some skilled people remove the bad programs.

193

Reply to this comment    5 March 2004, 12:54 GMT

Re: The Quality of Our Archives
Ayial  Account Info

I am definitely for choice 3. I never liked the popular vote on anything :) The archives take forever to find something in particular. I usually resort to doing a word search through the browser. Catagorizing should be done no matter what. It would just make it easier to find a file. I mean if there were a subcatagory for "economics" I would be so happy. I'd instantly write all these programs I've been meaning to upload.

Reply to this comment    5 March 2004, 14:13 GMT

Re: Re: The Quality of Our Archives
Sang Nguyen  Account Info

I agree because on 56K, it takes at least 2 minutes to load the TI-83 Plus BASIC Misc. section. Breaking it up into subcategories is a brilliant idea!

Reply to this comment    5 March 2004, 14:33 GMT

¤
burntfuse  Account Info

I totally agree. I have a dialup connection too, and even the 86 BASIC math programs page takes so long to load that I never visit it.

Reply to this comment    6 March 2004, 14:12 GMT


Re: Re: Re: The Quality of Our Archives
vadim p  Account Info

*gasp* did you try...ti 83+ basic game section? in cable it freezed for like 30 secs to load it!

Reply to this comment    6 March 2004, 23:57 GMT


Re: Re: The Quality of Our Archives
Ayial  Account Info

I've posted this in the survey, and I'm going to post it here too:
It's all about incentive. Me, being an economics major in college, if I had a subcategory labeled "economics" I would have tons of incentive about making economics programs. And I mean it goes all around. If you have a specialty, whether it's trig, calculus, or even accounting, and you have your own subfolder, it will encourage you to make high-quality programs because you know that only people interested in that area will go there and they will remember you because you helped them out. Who cares if the program is not super smooth or popular, if you have your own specialty, and a subcategory you can relate to, you will be proud of your little area of interest and make programs accordingly with pride. Clearly the only choice that will make people like me want to create and post more files is to make subcategories, NOT voting!!!!

Reply to this comment    6 March 2004, 03:46 GMT


Re: Re: Re: The Quality of Our Archives
ti_is_good_++  Account Info

Exactly.

Reply to this comment    7 March 2004, 06:07 GMT

Re: The Quality of Our Archives
kp Account Info

Maybe there should be started a new, clean archive. I think that would be better then sorting out 'bad' programs. So the realy good programs would be added either by voting or download stats or staff opinions or some like that, but it would create a archive containing all great and usefull programs. And the people will like that. Also 'subdirectoriing' is good to keep track of lots of programs, but maybe the subdir before the last supdir should be listed all on one page, because if folders have only e few programs it's realy hard to click and click to find a program.
So that was it, keep up the good work ticalc!

Reply to this comment    5 March 2004, 14:39 GMT

Re: The Quality of Our Archives
qbman  Account Info

I have an idea. This would be very quick, but may take a little time to implement.

Why don't we rank aurthors. Lets say that the lowest rank can upload a maximum of five programs. That arthor can chose to also have submissions removed to make room for updates. This will vastly limit the number of people who waste space by submitting a hundred useless programs, they will only be limited to five.

The problem with this is that some people have more than five good programs and wouldn't want to have to choose. So, back to ranking ideas, people would have to appeal to have their rank increased so they can upload more files. When an arthor appeals, his or her programs are then evaluated to see if the arthor has contributed quality material (for instance, having an average program rank of 3.5 of 5 (70%) or someghing).

Though the BASIC archives suffer the most from this problem, the ASM archives also have some useless programs or out of date versions that just take up space for no reason.

I think useing a ranking engine would be more helpful along with more subfolders. For instance, under games, there could be an action, strategy, classic, arcade, and many other categories of games to help speed up loading pages by making the average page shorter.

Message continued on another post.

Reply to this comment    5 March 2004, 15:35 GMT

Re: Re: The Quality of Our Archives
qbman  Account Info

Like I said, implementing this solution would be difficult, but what could be done is that an e-mail is sent to every aurthor that has more than five programs telling them to choose five to stay in the archives (the arthor should be able to request backups of the ones not chosen). Those who don't respond to the message wouldn't get to have their program kept (eliminating people who don't care anymore). After everybody has chosen their programs, then we could start the rankings and people can start appealing for an increase in rank.

I would be willing to wait the time necessary to carry out this kind of system. To make things more smoothly, the work might be broken down into sections (like work only on the TI-83 basic archives one week, then move to the 92 archives, one calc or major section at a time) This will allow most users to continue like normal untill their calc is the one being changed. The work would then be focused and trafic would still continue to the site.

(Correction to above post): First sentance should read:

I have an idea. This would be very quick and organized once in opperation, but may take quite a bit of time to implement.

Reply to this comment    5 March 2004, 15:39 GMT

¤
burntfuse  Account Info

That sounds like a great idea, except then even someone who was only allowed 5 programs could still upload a quadratic solver, so it wouldn't solve the repetetive program problem-still, a good idea.

Reply to this comment    6 March 2004, 14:15 GMT

Re: Re: Re: The Quality of Our Archives
Joey Gannon  Account Info
(Web Page)

Of course, if I had only been allowed to upload five programs, I would probably not be where I am today. I'm not really sure whether that's a good thing or not. :-)

Reply to this comment    7 March 2004, 05:04 GMT


Re: Re: Re: Re: The Quality of Our Archives
BlackThunder  Account Info
(Web Page)

I have a feeling someone will flame you, then realize you're a filearchiver. :)

Reply to this comment    9 March 2004, 04:15 GMT

Re: Re: Re: The Quality of Our Archives
Eugene Talagrand Account Info
(Web Page)

Most of the programs people object to in the archives (BASIC shells) usually are written by people who only submit one file.
And many people who make this a fulltime hobby have more than 5 excellent programs in the archives.
Also, not to sound too grim, but some authors have passed on and can't exactly update their files.

Reply to this comment    8 March 2004, 07:32 GMT


Re: Re: Re: The Quality of Our Archives
Ben Cherry  Account Info
(Web Page)

so if you emailed Bill Nagel, and he didnt respond then it would be okay to delete Mario 86? I think not. I think Morgan thinks not too ;)

Reply to this comment    8 March 2004, 17:55 GMT


Re: Re: Re: Re: The Quality of Our Archives
qbman  Account Info

Hmm... I did overlook the fact of the people who no longer program. Maybe a prerating system (rating before doing the rest) would allow people to vote to keep their favorite program (and could give some people starting rank if many of their programs recieve good votes). Programs rated poorly and/or programs written by aurthors who did not respond or recive votes could be sent to a folder for unpopular or out of date programs.

I've read a majority of all the posts here, (but not all, over 400 now). Obviously, the file archivers don't want to delete anything (they probably don't want to make anyone feel bad), so organization becomes the only way to solve the problem (I say that this problem can't be just overlooked). Maybe our problem's solution should be broken down into pros and cons like so...

Pros of deleting junk
-less files have to go through to find something
-less of a chance of encountering a worthless program
-faster loading times all around

Cons of deleting junk
-intimidating to newbies
-discourages aurthors to have their program labeled trash and treated as such
-more work for the archivers

If people would put thier ideas to this kind of test, it would help to quickly realize which Ideas are best. I would make a chart for each of my ideas, but I don't have the time now, so I welcome others to create pros/cons charts for not only my ideas, but for others' ideas as well.

BTW, I'm not a file archiver, nor do I have the time to be one, but I really do appreciate the work those guys (or gals) do.

Reply to this comment    9 March 2004, 17:32 GMT


Re: Re: The Quality of Our Archives
Professor Account Info

Would you idea allow the newbie who want to post go up in ranks if their program became well liked?

I am about to post one of two year-long endeavors.

I have yet to ever post a program here.

Reply to this comment    5 March 2004, 15:42 GMT


Re: Re: Re: The Quality of Our Archives
qbman  Account Info

If the program became well liked, then their program would have a high enough rating to help them out, but to go up in rank, that person should have at least five programs (no need to be able to post 100 programs when you only have one program)

My idea is to encourage newbies. A ranking system would encourage people to increase ranks to look better, but to do that, they have to make good programs that people like. Going up in rank is not a matter of number of subbmissions or time a member, going up in rank should be based on quality of submissions, so back to your question: A newbie with 5 well liked programs that have a good enough average rank will be able to appeal to get an increase in rank.

Something more I thought of: Though going up in rank could be done automatically, a committee should review people wanting to go up in rank so that aurthors couldn't have friends vote 5's on bad programs and get rank increases easily. When going up in rank, the committee would only have to consider the limited submissions of the aurthor and the average ranking of the submissions. Since a computer can't judge quality, I am very for a committee; in fact, I would be happy to help out on a committee for this.

Reply to this comment    5 March 2004, 15:51 GMT


Re: Re: Re: Re: The Quality of Our Archives
Professor Account Info

Thanks for your input.

I just think appeals should be bypassed and the rating of the 5 programs will automatically raise the rank of the contributor.

Reply to this comment    5 March 2004, 19:40 GMT


Re: Re: Re: Re: Re: The Quality of Our Archives
Ben Cherry  Account Info
(Web Page)

I disagree completely, i dont think that authors should be ranked like that. That is unfair to the authors who have a large number of useful, yet simple programs designed for a certain niche. And in general it is just unfair and intimidating to all authors.

Reply to this comment    6 March 2004, 06:11 GMT


Re: Re: Re: Re: Re: Re: The Quality of Our Archives
qbman  Account Info

The only time the 5 program limit would be very restricting is while the archives are being rebuilt, but as soon as they are, people would then be able to go up in rank and be able to post the rest of their programs. If they truly have useful programs, then they will have high ratings that encourage the committee to raise the rank of the aurthor to allow more programs.

The solustion repetetive programs can only be filtered out useing a committee; besides, the popularity of a type of program that has already been made will be less for each person's version after the first (assuming all have the same feature). Though ranking doesn't eliminate the problem directly, people will learn that if they want to go up in rank, they will have to make something worthwhile and different. What I'm saying is that if people know that a pong or quadratic solver isn't going to help their rank, then they might not want to make one unless they think they can make one better than the currently existing most popular version. Does this make sense?

I mentioned earlier that an average would be a consideration for advancing in rank. At first, when the archives have been rebuilt, not all programs will have been voted for, so if an aurthor feels they have quality programs, but people haven't voted yet, then the committee can advance an aurthor's rank purely based on quality (a choice no computer is able to make). This would also give the ranking system a more realistic feel so people know that if someone with a bad attitude went through and voted 1's on all a particular aurthor's programs, the committee could still pass them on if their quality is good.

My main reason for wanting a committee is that a computer is a good judge of quantity, but a poor judge of quality; however, a human may be average at judgeing quantity (slower counting and bad at math), but is many times better at judgeing quality.

Reply to this comment    6 March 2004, 20:11 GMT


Re: Re: Re: Re: Re: Re: Re: The Quality of Our Archives
Ben Cherry  Account Info
(Web Page)

I understand your system, but i think that the ranking system would still be unfair and discouraging. Maybe it would work, but i dont think that it would work out right. And what committee is going to look over all of the authors. Look at the userid numbers on some people, like myself. Im about 45000 and my account was created 6 months ago. about half of the accounts are authors i think, so that gives 20000 to 25000 different authors. A committee would have to look at each of those once at the creation of this system before the computer can be given free reign over the rating.

Reply to this comment    6 March 2004, 21:12 GMT


Re: Re: Re: Re: Re: Re: Re: Re: The Quality of Our Archives
jrock7286  Account Info

sift through the authors list...search for an author named "asdfasdfasdf"...sadly, you WILL find one. This upsets me a great deal.

Reply to this comment    7 March 2004, 06:08 GMT

1  2  3  4  5  6  7  8  9  10  11  12  13  14  15  16  

You can change the number of comments per page in Account Preferences.

  Copyright © 1996-2012, the ticalc.org project. All rights reserved. | Contact Us | Disclaimer