Add users feedback ratings for LibStock Downloads?

[Megahurts]

Hi,

I am wondering if others would also like this change to LibStock downloads?

Currently, the "Star" rating system is based on whether the author used the "Package Manager" software to bundle their offering or not it seems, to me anyway.

My first upload to LibStock (Random Team Draw), was done using it, and instantly there was 5 stars showing in the ratings.
I was not even sure I used the Package Manager software correctly to make the package, but since no one reported anything wrong, I assumed I did not do anything terribly wrong, but worried I missed setting something in it right, Still.

I do not think/feel that system is a fair and accurate method for people browsing thru LibStock looking at the offerings to help them determine if a offering or another would be the better one to download and try.

How about making the Star ratings be user feedback driven based on their evaluations of a few categories like:

Functionality - It actually does what it was made to do and package (or Zip) was complete, nothing missing from declared content. (Low)1--to--5(High)

Usefulness to the User: 1 - - - 5

Directions for Use: 1 - - - 5

Ease of Use: 1 - - - 5

Author Support: 1 - - - 5


And/or any other categories anyone would like added or used instead.

I feel everyone that uses the site would find something like this more helpful with finding the right download they wanted/needed
and the authors would then actually have some real indications of their efforts or that attention is required.

Opinions anyone?

With best intentions and regards, Robert.

[Toley]

Hello Robert I have asked for something like that for the forum posts and look how I've been crucified!

http://www.mikroe.com/forum/viewtopic.php?f=146&t=59203

[JohnCode]

Hi Robert

for me this is a good idea. Also people should write at least a sentence. I 've got a project and received one star. I can accept that, but I would like to know what is so bad. So I can learn something and improve my project and code. But without any feedback I don't know why I get this rating, so I can't change anything.
Furthermore there are some really good projects, but they don't have any documentation, so it's hard to use them. So i think it's a really good idea to rate like you proposed:

Usefulness to the User: 1 - - - 5

Directions for Use: 1 - - - 5

Ease of Use: 1 - - - 5

Author Support: 1 - - - 5

[Dany]

Perhaps an observation on the current star rating system (as far as I know and can observe): the number of voters is not shown, and the rating shown is the average of all votes.
This means that 1 vote of 5 starts gives also a rating of also 5 stars, which is of course a little bit misleading (or at least not very informative).

Perhaps it would be better (again referring to the current system)
- only showing rating if the #voters e.g. is >= 10.
- also show the #voters.

[janni]

I don't think people will be willing to take part in any detailed rating (and probably not many are aware that they may actually rate projects). Nobody stops people from writing comments but, judging by the small number of comments and their contents, people write something almost exclusively when they have problems or spot errors - in other words, when they need something. That's nothing strange, of course, Libstock isn't a social network, people search for solutions and do not come back if everything works (note the very small number of users following even most popular projects).

While it's clear that detailed rating would be a big help in choosing projects, I wouldn't count on Libstock users changing their behaviour - even if it could help them in future.

Dany's idea to have present rating made more reliable by adding number of voters is the best we may hope for, as it could be done by mE.
At the moment the only reliable rating is the number of downloads, though it mostly reflects popularity of the subject the project is dedicated to, not the quality of the project itself.

[Megahurts]

Hi Guys,

Thanks for pitching in on this and pointing out the shortcomings (and of current system).
(still waiting for official response(s) from mE. staff though)

1st @Toley: (sorry for this late reply) I did see that before posting this one and the treatment you got got almost made me not do this try at "changing the (current) system" (to fix it I feel).
But I finally decided to try it anyway, because they are very different requests after all. I have no opinion to say on where I would stand on your request either,
as it was a request to change or add something to the forums features, and not to fix something present in the system that had issues being used/abused or implemented.
So since it does not exist, I really would not know how to 'feel' about that topic of debate. Sorry you got such inconsiderate tactless responses too.
(I do 'feel', that was uncalled for)

2nd, on topic:
Agreed Dany, if only one users rating has been submitted, it should indicate such clearly. But after that, the average of each category
rating should be done and shown as statistical averages, with each users rating submitted (with comments) able to be seen by all.
I think that should be a fair indicator users can then use to guide them to what they are looking for, and hopefully (be used to) provide the person(s)
that submitted the upload the feedback they need, to know if anything needs fixing or everyone is happy with the submission.

If a 5 star rating (now) is wholly dependent on whether the Package Manager software was used to bundle the submission or that its use or not
has any effect on what a submissions rating is, then the rating system is worthless because it is then 100% biased or misleading to
the users and provides no worthwhile function or purpose.

Personally, I do not like using the Package Manager software for many reasons, and do not think it should be used by anyone to prepare
anything for submissions unless the submission requires it be used because of the nature the submission is, IE.. a Library or application
that requires installation of files that the compiler will need to work with the application submitted.
Otherwise, it seems that the Package Manager software has too much potential and capability to do damage to a users installations of mE software.
It is far to complex to use for submissions of simple applications, projects or examples and not needed for such type submissions there.

I do not know for certain, because I only used it one time, but after 'playing' around with it trying to make a package of an example project I made,
I was not comfortable with seeing what crucial compiler files a inexperienced user of it could easily effect unintentional changes or damages to.
I can see its need for certain content, but for general use of packaging most submissions, its use makes me wonder how many problems
forum users report with their HW/SW not working as advertised are actually caused by a bad package being gotten from LibStock and installed?

So if it is the source of a LibStock submissions rating in whole (need a response on this please mE staff) and planned to be the only criteria
for the rating system, I then think the whole rating system is worthless and needs removed entirely.
Even a bad package uploaded would get 5 stars and stay that way as there is no way a user can provide any feedback that would affect it.
So until ratings system is changed to user feedback driven only, I feel it is a worthless indicator for guidance and has no value or merit being displayed at all.

(edit): @janni (you got yours posted before I completed mine). Spot on there also and I agree totally. It would only work if used and maybe
mE. did it this way because no one used it before, but that does not correct it or make it of any value implemented like it is now.
I am guilty of not doing surveys or going back to a site to give feedback also, maybe 1% of a lifetimes of downloads have I done it on.
Some (most) I forgot where I got it from by the time I get around and actually try it (the download), some I will not jump thru the processes required to submit one.
Very few places (sites) have a system that nurtures it being used and makes it as easy to participate as possible for the public.
I don't do 'Social Networks' anymore also. The structure and nature of them are a waste of my time and I do not want my PC polluted with the resources
they need, my PCs work for me when I demand it, not the other way around or for them (Snetworks).

(and probably not many are aware that they may actually rate projects)

I for one am not aware of anyway to rate a LibStock offering! How does one do that?


Fair is fair, wrong is wrong, everyone knows this, or should.

[Jim Walsh]

Hi,

To rate the projects on Libstock, you have to roll your cursor over the stars in the ratings line when you are you on the project's page, then click on how many stars you want to give it.

Peace,
Jim

[janni]

Personally, I do not like using the Package Manager software for many reasons, and do not think it should be used by anyone to prepare anything for submissions unless the submission requires it be used because of the nature the submission is, IE.. a Library or application
that requires installation of files that the compiler will need to work with the application submitted.
Otherwise, it seems that the Package Manager software has too much potential and capability to do damage to a users installations of mE software.

You're right, use of Package Manager makes sense only for libraries. As for any danger of damage, I wouldn't worry that much - it may be easily limited to one processor definition files and library installation is reversible.

So if it is the source of a LibStock submissions rating in whole (need a response on this please mE staff) and planned to be the only criteria for the rating system, I then think the whole rating system is worthless and needs removed entirely.
Even a bad package uploaded would get 5 stars and stay that way as there is no way a user can provide any feedback that would affect it.
So until ratings system is changed to user feedback driven only, I feel it is a worthless indicator for guidance and has no value or merit being displayed at all.

mE has naturally the right to rate a project (and other users may always outvote it) but I do not think use of Package Manager is the sole criterion in their ratings.
At the moment rating is indeed rather worthless indicator of project quality and gives little guidance for anyone. As always with statistics, pure numbers without context tend to mislead instead of help. Only after implementing Dany's suggestion the ratings could get some meaning (the one about # of voters - showing only ratings with more than 10 voters would probably remove all ratings ).

I am guilty of not doing surveys or going back to a site to give feedback also, maybe 1% of a lifetimes of downloads have I done it on.
Some (most) I forgot where I got it from by the time I get around and actually try it (the download), some I will not jump thru the processes required to submit one.

We may naturally appeal to all Libstock users for more active participation in the rating system, but most will simply 'do the shopping' and never go back to the same project. And any attempt at forcing users to perform detailed rating while voting will further decrease the number of votes.

To rate the projects on Libstock, you have to roll your cursor over the stars in the ratings line when you are you on the project's page, then click on how many stars you want to give it.

One may also rate comments in the same way. Fortunately, one cannot rate own project/comment .

[Megahurts]

Thanks Jim & Janni,

I did not know that about the ratings being visitor adjustable, or that you cannot change your own. That's too bad, I would really like to adjust
the one I used the package manager on down to 1 star. It sure didn't deserve 5 stars for the effort and functionality I put into it.
Maybe I'll ask for some friends to go slam it down for me then eh?
(It is the ONLY one with rating stars.......I'll turn my back for a few.....)


So,... does the restriction still apply if you are not logged in then? Or do users have to be logged in to make any ratings entry?

If there was at least some clear (and visible) indicators or clues or something that visitors could/would see that let them know a rating submission could be given, there might be more users doing so? maybe?

If it turns out that I was the only one that didn't figure it out, I retract my rantings and enter a plea of insanity, due to too many squirrels distracting me. whaat.. squirrel ... where?

I agree that any system that requires or interrupts a visitor to take or do a rating survey is the last thing I would want to see done.
I hate those schemes the most. How can somebody rate something if they have not used it yet, let alone downloaded it at some places

Starting to think that computers (good idea) made us a little smarter, but the internet (still debating on it) has made us a lot dumber.
Wow, I was starting to miss AOL there for a fraction of a plank time.

Well, if people won't use a rating system if given the opportunity, and it appears that we were given it, (yeah, making a 10 rating
minimum probably would clear out the ratings @janni & Dany), then I beg forgiveness and drop my case so we can do something
better with our time,.... like chase some squirrels......
Arf!
(this should teach me to think out loud, again...but probably not )

But I will go check out the ratings thing @LibStock to see if there are any squirrels there / I mean something to click on, like a few stars
I should have given to some gracious authors for their work and efforts (Dany, Janni, Aleksandar to name a few ).
(and sorry , had I known , I would have used the rating system there, I know how it is to get some positive feedback for your efforts Next best thing to getting paid for it right )

Thanks again for the info guys, and for your thoughts, valuable as always, Robert.

[marina.petrovic]

Hi,

Thank you all for your suggestions, we really appreciate it.
I will pass this forum topic to our web admins, and they will certainly consider all suggestions and remarks.

Best regards,
Marina

[janni]

Thanks, Marina .

So you see, Robert, your efforts weren't in vain .

So,... does the restriction still apply if you are not logged in then? Or do users have to be logged in to make any ratings entry?

Only logged users can vote, which partly explains low activity in this respect.

If there was at least some clear (and visible) indicators or clues or something that visitors could/would see that let them know a rating submission could be given, there might be more users doing so? maybe?

Probably. I discovered the possibility accidentally (must have been looking for squirrels ). Many may have less luck.

[Megahurts]

That was fast replies this time, only got two other posts done, wait... ok, maybe not as fast as I first thought.
One post was short, the other was in keeping with my habit of typing until the coffee wanted back out or there was a squirrel needing chased from my apple tree.
(its our Pomeranian's fault I chase squirrels now, also. Others may call it walking the dog, but not in my/our case, its a race! )

Thank you very much for letting me (us) know you guys seen this and will take into consideration Marina. Love ya and glad you started working there (and here in the forums ).

Only logged users can vote, which partly explains low activity in this respect.

Bingo!

(must have been looking for squirrels ). Many may have less luck.

You Too!
So maybe adding a few hidden squirrels linking to ratings usage info to find like Easter eggs for everyone will solve this?
(never thought anything good would come from squirrels though )

ps. I actually like the little fury critters, keeps me and the dog in shape and senses sharp.

Thanks again for the discussion everyone, Gotta Go, Squirrel...........