User-contributed timetables - a fresh approach needed?

Geoff, for crying out loud, all I'm suggesting is that timetables are tested before release. The sky isn't falling in and it doesn't mean the end of Simsig. The aim is to improve the product - that's all. Nothing more, nothing less. How its done does not bother me one bit and all I was doing was suggesting ways this may be achieved. It doesn't mean that the control of the way Simsig is going changes in any way. You're still in control

and and well.....just what is all that psyco-babble about in Point 1?

As for Point 5 I don't have factual evidence and neither do you. You're going on your gut feeling just like me.

Interesting. I have an Exeter Timetable which is being tested by a small goup of people. The data is taken from cif / TRUST / Engineering Haulage / OTM sheets. The scenario is based on last September and there was a week long Blockade between Taunton and Exeter ( I can give the precise posession points if required). The data produced an alarming amount of schedule clashes.
If this Timetable went before a "Timetable disciplinary Panel"
1. Would it be classed as a bad timetable due to the clashes.
2. Should I replatform trains in the timetable to get rid of the clashes
3. Should I leave the clashes in and let the "Bobby" sort it (As in real life).

How would this be rated?

Chrisrail in post 134650 said:

Interesting. I have an Exeter Timetable which is being tested by a small goup of people. The data is taken from cif / TRUST / Engineering Haulage / OTM sheets. The scenario is based on last September and there was a week long Blockade between Taunton and Exeter ( I can give the precise posession points if required). The data produced an alarming amount of schedule clashes.
If this Timetable went before a "Timetable disciplinary Panel"
1. Would it be classed as a bad timetable due to the clashes.
2. Should I replatform trains in the timetable to get rid of the clashes
3. Should I leave the clashes in and let the "Bobby" sort it (As in real life).

How would this be rated?

That has certainly been the dilemma timetable authors have debated since forever. You'll never get a consensus. I have noticed some authors do write comments like "platform clashes left in" which, I guess, might alleviate any doubt, RTFM issues aside.

As time goes on though, how much of what you see today is loader and sim changes breaking timetables that once worked ?
I currently have 2 of mine deleted (Motherwell) and 2 'less than ideal' (CSCOT/COW 84) and have only just repaired Motherwell 93 due to Loader alterations.

bill_gensheet in post 134599 said:

As time goes on though, how much of what you see today is loader and sim changes breaking timetables that once worked ?
I currently have 2 of mine deleted (Motherwell) and 2 'less than ideal' (CSCOT/COW 84) and have only just repaired Motherwell 93 due to Loader alterations.

What were the loader and sim changes that broke things and how much warning did you get? As I've said before, loader changes shouldn't break things.

As a brief note of clarification there are a lot of older TTs in the Downloads section which are shown as being submitted by me. This is an accident of history from when we moved from the old web-site to this one. IIRC there was only about 48 hours to complete the move and you were under work pressure and short of time so I moved the old TTs across. However, the only way I could complete the moves was to submit them in my name. Where the information was available the file details show the original author if you wish to keep that person in the loop.

Interesting. I have an Exeter Timetable which is being tested by a small goup of people. The data is taken from cif / TRUST / Engineering Haulage / OTM sheets. The scenario is based on last September and there was a week long Blockade between Taunton and Exeter ( I can give the precise posession points if required). The data produced an alarming amount of schedule clashes.
If this Timetable went before a "Timetable disciplinary Panel"
1. Would it be classed as a bad timetable due to the clashes.
2. Should I replatform trains in the timetable to get rid of the clashes
3. Should I leave the clashes in and let the "Bobby" sort it (As in real life).

How would this be rated?

So here is a draft proposal. A lot of people have said a lot of things so I may not have captured it all.

1. New TT submissions will be listed on the sim product page instead of a separate file area. Initially they will be categorised as "under test".
2. An approved user, or users, will test the timetable and approve it for use. This will move it to a "released" state.
3. At any point, any registered user can add a review with stars.
4. Pertinent information about the file will be more obvious - upload date, user, etc.

In the meantime, files that are in the Downloads section will not automatically transfer over. The original author, if still around, can elect to have it moved (or re-uploaded). In fact, any user should be able to do this. It should still be at stage 1 though.

We do need to keep track of when a file is under the product page so it can be removed from the Downloads area.

Any files left after some period of time can be put in some archive - still accessible, but "enter at your own risk". Similarly, timetables found not to work (eg because the sim has changed too much) can go into a "non-working" archive where it can stay forevermore.

Te "approved" user in #2 would ideally be a large set of users from here. Verifying old uploads is quite a big task so it needs a lot of people to volunteer.

These steps will also help to integrate the Loader with user-contributed content - for example, on the Loader, when selecting a timetable, an entry could be "click for user-contributions" which will pop-up a window where the user can see the available timetables for that simulation. The user can pick which to download while being shown a description, version, author, ratings, reviews. It will also automatically install itself in the correct file location. Again, the user would be in control of this.

So here is a draft proposal. A lot of people have said a lot of things so I may not have captured it all.

1. New TT submissions will be listed on the sim product page instead of a separate file area. Initially they will be categorised as "under test".
2. An approved user, or users, will test the timetable and approve it for use. This will move it to a "released" state.
3. At any point, any registered user can add a review with stars.
4. Pertinent information about the file will be more obvious - upload date, user, etc.

In the meantime, files that are in the Downloads section will not automatically transfer over. The original author, if still around, can elect to have it moved (or re-uploaded). In fact, any user should be able to do this. It should still be at stage 1 though.

We do need to keep track of when a file is under the product page so it can be removed from the Downloads area.

Any files left after some period of time can be put in some archive - still accessible, but "enter at your own risk". Similarly, timetables found not to work (eg because the sim has changed too much) can go into a "non-working" archive where it can stay forevermore.

Te "approved" user in #2 would ideally be a large set of users from here. Verifying old uploads is quite a big task so it needs a lot of people to volunteer.

These steps will also help to integrate the Loader with user-contributed content - for example, on the Loader, when selecting a timetable, an entry could be "click for user-contributions" which will pop-up a window where the user can see the available timetables for that simulation. The user can pick which to download while being shown a description, version, author, ratings, reviews. It will also automatically install itself in the correct file location. Again, the user would be in control of this.

May I suggest 'Not tested' and 'Tested' for 1 and 2, as opposed to 'Under test' (some may not be, or at least not for some time) and 'Released'? Just makes it a little more open, a little more welcoming for writers who want to play and who have no particular wish to get their products rigorously tested.

So here is a draft proposal. A lot of people have said a lot of things so I may not have captured it all.

1. New TT submissions will be listed on the sim product page instead of a separate file area. Initially they will be categorised as "under test".
2. An approved user, or users, will test the timetable and approve it for use. This will move it to a "released" state.
3. At any point, any registered user can add a review with stars.
4. Pertinent information about the file will be more obvious - upload date, user, etc.

In the meantime, files that are in the Downloads section will not automatically transfer over. The original author, if still around, can elect to have it moved (or re-uploaded). In fact, any user should be able to do this. It should still be at stage 1 though.

We do need to keep track of when a file is under the product page so it can be removed from the Downloads area.

Any files left after some period of time can be put in some archive - still accessible, but "enter at your own risk". Similarly, timetables found not to work (eg because the sim has changed too much) can go into a "non-working" archive where it can stay forevermore.

Te "approved" user in #2 would ideally be a large set of users from here. Verifying old uploads is quite a big task so it needs a lot of people to volunteer.

These steps will also help to integrate the Loader with user-contributed content - for example, on the Loader, when selecting a timetable, an entry could be "click for user-contributions" which will pop-up a window where the user can see the available timetables for that simulation. The user can pick which to download while being shown a description, version, author, ratings, reviews. It will also automatically install itself in the correct file location. Again, the user would be in control of this.

3. What was the criteria for checking by the testers and was this relaid to the author.

I do agree that some of the submiitted timetables leave things to be desired and something could be done about it before the proposed timetable is even submitted for inclusion in the suite for each sim.
I have previously been involved as a tester of timetables for another competitors submissions and feel that the pitfalls encountered should at least be passed on before we take this task as a given.

1. One of my main bug bears was that you actually had to own the sim that the timetable was for. No temporary access was granted for testing purposes and would Geoff be willing to let people have access which would then not only be used for testing. Sims cost money. The bigger the sim the more money needed to buy it. This consequently led to me thinking that it was becoming a expensive way to test someone else’s work and did I need to do this.
2. How are the testers going to give feedback on the errors that they might find. In my experience it was drip fed back to the author and they the author sometimes did not implement all the changes requested. This led to a long process at times and in some cases the author became discouraged and were never heard of again leaving the timetable in mid air and never to see the light of day.
3. What was the criteria for checking by the testers and was this relaid to the author. Speed of run through, timetable notes were the general rule of thumb for errors to be noted and passed back.
4. Speaking as someone who likes to play around creating timetables, I have never fully understood everything that needs to be written into one before offering it up. Things like decisions, rules etc are not my best or easiest things to understand but fully appreciate it comes as second nature to some. Guidance needs to be given and written for people to fully understand what will be taken into account when it is being tested. The wiki does not and cannot give all of the requirements and how they are implemented.
5. Some people who submit timetables have preferences for certain eras I.e 70’s, 80’s and 90’s. Would they only be interested in checking timetables of that time or would they be prepared to test those of today’s offering.
6. Surely it must be understood that some people cannot possibly check massive sims like Three Bridges or those that do not have ARS to help by themselves. We all know that ARS is not fool proof and does have it’s drawbacks.
7. Who is responsible for the changes that may be required before acceptance.
8. Is this exercise going to be to the detriment of new simulations not being worked on and resources being diverted else where. Some who have been patiently waiting for a new release for some time may see this all as an unnecessary diversion.
9. No one is perfect and errors will creep in so please do not push new users away by making life to difficult.

Hope the above is useful before we proceed with this.

Geoff

I do agree that some of the submiitted timetables leave things to be desired and something could be done about it before the proposed timetable is even submitted for inclusion in the suite for each sim.
I have previously been involved as a tester of timetables for another competitors submissions and feel that the pitfalls encountered should at least be passed on before we take this task as a given.

1. One of my main bug bears was that you actually had to own the sim that the timetable was for. No temporary access was granted for testing purposes and would Geoff be willing to let people have access which would then not only be used for testing. Sims cost money. The bigger the sim the more money needed to buy it. This consequently led to me thinking that it was becoming a expensive way to test someone else’s work and did I need to do this.
2. How are the testers going to give feedback on the errors that they might find. In my experience it was drip fed back to the author and they the author sometimes did not implement all the changes requested. This led to a long process at times and in some cases the author became discouraged and were never heard of again leaving the timetable in mid air and never to see the light of day.
3. What was the criteria for checking by the testers and was this relaid to the author. Speed of run through, timetable notes were the general rule of thumb for errors to be noted and passed back.
4. Speaking as someone who likes to play around creating timetables, I have never fully understood everything that needs to be written into one before offering it up. Things like decisions, rules etc are not my best or easiest things to understand but fully appreciate it comes as second nature to some. Guidance needs to be given and written for people to fully understand what will be taken into account when it is being tested. The wiki does not and cannot give all of the requirements and how they are implemented.
5. Some people who submit timetables have preferences for certain eras I.e 70’s, 80’s and 90’s. Would they only be interested in checking timetables of that time or would they be prepared to test those of today’s offering.
6. Surely it must be understood that some people cannot possibly check massive sims like Three Bridges or those that do not have ARS to help by themselves. We all know that ARS is not fool proof and does have it’s drawbacks.
7. Who is responsible for the changes that may be required before acceptance.
8. Is this exercise going to be to the detriment of new simulations not being worked on and resources being diverted else where. Some who have been patiently waiting for a new release for some time may see this all as an unnecessary diversion.
9. No one is perfect and errors will creep in so please do not push new users away by making life to difficult.

Hope the above is useful before we proceed with this.

Geoff

If a writer wants to offer an early attempt that's known to have lots of faults in order that they may benefit from feedback, I see no reason to stop them. I see every reason to warn potential users what they're letting themselves in for (and some, of course, will delight in working out how to overcome the problems, and will be generous with feedback that will help the writer develop).

Sim ownership will always be necessary. Not all testers will be able to work on all timetables.

Some more thoughts, some of which address points raised recently.

1. I don't think anonymisation for the purposes of testing is going to work too well. There aren't enough authors so it would probably be quite easy to work out who an author is. In addition, the timetable should be complete, so credits and copyrights should already be there - the copyright generally being the author's name. I do agree with the idea; just not at this time.

2. Timetables should not be held back pending testing. People spend time writing and testing them, and it's unfair to have to wait for an indeterminate length of time for it to complete testing.

3. Rather than give a "untested" or "tested" status to a timetable, perhaps a more positive approach is simply that tested timetables get a "certified" banner affixed to it - ie blank for untested. Avoid negativity.

4. Tester for these timetables would not be the testers for the simulations, or those that test "official" timetables. I would expect a bunch of volunteers to handle it.

5. You would need to own the sim (or the sim be a free one) to test on it. There are plenty of people available for each sim without giving away even temporary licenses. I would also argue that temporary licenses would be detrimental to the testing process as the user is less likely to be familiar with the area.

I've read everything that's been written so if I haven't addressed it, it's probably because I don't have a strong opinion either way on it! But ultimately, as these are community timetables, perhaps the community should select somebody to lead it and I'll support in any way I can.