Schrodinger's Cat Strikes Back

Home » General discussion

Category Archives: General discussion

PhysicsOverflow at the 5th Offtopicarium

Since Wednesday lat evening, I am back happy and still a bit tired from the 5th Offtopicarium in Wegierska Gorka, Poland.  As the Offtopicarium is organized by science nerds for nerds ;-), it was a lot of fun to be together with the very nice small (about 30 people) group of young people. The program was rather interdisciplinary, ranging from biology, history, constructed languages, space projects of students, science communication and education, and many more things. I hope all talks will be accessible soon from the official homepage. In particular the talk about Opening Science  could be worthwhile to consider in some detail, as there is some overlap to what we want to achieve by the Reviews section. Getting to Wegierska Gorka from Bad Doberan (Germany) was a bit cumbersome and time-consuming.  From Katowice I still had to take another train for 2 hours. Considering different means of public transportation, proper time of the trip can not be reduced to less than 14h+  which is therefore the global minimum. route Therefore I was not able to arrive before Saturday afternoon. The meeting had begun on Thursday already and missed therefore about the first half of the talks. All talks have been very good and impressive generally, here you can see  for example Piotr Migdal giving a talk about how often the best and most innovative things are done by people with greatest passion and motivation in their free time, even though (or because) they do not get paid for it. As PhysicsOverflow for example …  😉 IMAG0284 My talk about PhysicsOverflow Happily, my talk about PhysicsOverflow was soon enough ready as you can see below. And surprisingly I realised that I did not even have to take notes about what I’m going to say (as for other talks related to my work) but I could just go on blathering  without effort … ! When bragging about PhysicsOverflow, I am obviously unstoppable 😛 Unfortunately, my laptop refused to connect to the WLAN in the meeting room, so I could not show a live demonstration :-/. You can find the slides of my talk here (the quality was better in PPT than it now is in PDF)  🙂 IMAG0290 In the introduction, I first explained why a new higher-level physics site was urgently needed, how PhysicsOverflow is organised in different parts such as Reviews, Q&A, Meta, etc as well as our most important principles and characteristics. On the next slides, I introduced the Q&A part as a physics analog of MathOverflow and explained the most basic things about how it works, scope, etc … To introduce the Reviews section I ranted a bit about how present journal peer-reviewing sucks ;-), then explained how paper reviewing is done on PhysicsOverflow, and showed two examples of a negatively and a positively reviewed paper. I also got a very good discussion to my talk (nicely supported by Piotr Migdal) , the following issues have been discussed (not in exactly the order listed here):

  • Why did we have to creat something new instead of using something that already exists, such as for example the SE platform?

To answer this, I had prepared in advance a slide (22) that summarises a bit why the goals of the SE company IMO clash with the intentions of a high-level academic community. Somebody mentioned that there are academic sites in the network too which is true but works only  for exceptional cases (I only know MO and CST) … Piotr helped by explaining that very high-level questions are not really high-lighted on SE for example. Also, conversely to the time when MathOverflow was created almost exactly 5 years ago (congratulations!) , SE does no longer give away its software today. There have also been

  • Some discussion about partitioning the site instead of supporting other communities in starting their own site (for example PhysicsUnderflow) to build an Overflow network

This is in principle doable as we have categories in addition to tags,  as for example explained on slide 20  about the differences between the PO and SE software.

  • Of course, I also called for a second developer

which resulted in the suggestion to upload the code on Github for other people to help us with the development and debugging, but as Polarkernel said we are not yet ready to do this … There have also been some

  • Questions about how many people we have, how many papers are submitted, etc :

About 250 newly registered, not all are active at the same time of course … The Reviews section (Reviews I) went online later than the Q&A section, so it is natural for it to have less content  (48 papers at present) than the Q&A section. Another questions was about

  • What could endanger the success of the site?

Here I said that at  the beginning, I and probably others were too enthusiastic about the Reviews section finally getting started, so we were too permissive about accepting submissions. But this is fixed now by our “reject to review policy”. There was also

  • Some discussions about the formula to calculate the final score of submissions:

Strong nonlinearity, the effect of additional points for one unit  accuracy and originality depends in  not plain obvious way on the votes the paper already has… As the accuracy determines the sign of the final score, crackpots who make unfounded “revolutionary” claims are punished which is a good feature. Does this formula really what we want to achieve in all cases? Final remarks The meeting was really fun, and there have been many cool nice people  :-). I am very happy about the positive response I got to my introduction of PhysicsOverflow and will also reconsider some of the other great talks I have seen. And I really liked it to meet Piotr Migdal in person 🙂 BTW he has now submitted his PhD thesis, congratulations ! Only the mud (literally!) sucked on our trip to the mountains because it was rainy the days before …  😀 P1110416 Joking … The trip was fun too 🙂

A long step towards the adult PhysicsOverflow

Today, the new category system, which allows an unrestricted number of category-levels, has been installed on PhysicsOverflow. This system was required to enable the categorization of submissions beyond the four levels provided by the actual Question2Answer (Q2A) framework. This new system is almost invisible for the user, but has the important function to make a (future) large number of submissions searchable by the user. About two and a half months have been required to develop this system, you had to wait for a long time. Therefore I like to give you some insights.

The original category system of the Q2A framework is a hierarchical database model. Every node (a category) is linked to its parent by the ID of the parent node. Such a model enables a fast writing of new nodes (you just create the node and link it to its parent), while queries through the tree are usually slow. Q2A solved this issue by a fixed number of 4 category levels. To be able to find parents in a short time, the path to the parent is hard-coded in every post. This means that every post contains four indices storing the way back through the tree. This is a clever and fast solution. However, it can not be extended to an unrestricted number of categories, it even gets slow if extended to 8 levels for instance. The category system is written in the core code of Q2A and spread over a large part of the system.

Another issue arising when we increase the number of category levels is the user interface. Actually, when the user asks a question, he has to select the category for this post. The user interface uses select tags like:

categories_old

It is clear that for instance for eight categories, the place on the page is too small to display all these tags side by side. Therefore also the user interface had to be changed. The new user interface take much less room and looks like this:

categroies_new

For the new category system, I have implemented another database model, called nested set model. It allows for an unrestricted number of category levels. While queries through the categories become very fast, the insertion of nodes is slow, because all indices of the whole tree have to be changed. However, changes on the categories will be used much less frequent than for instance to display the tree. Like this, it is well adapted to our needs.

As already mentioned, the original category system is placed in the core code of Q2A and is active in almost all pages provided by the framework (even in pages I never expected this). I had to replace 9 files completely, all select specifications for database accesses to the category system had to be renewed and, naturally, the code for the nested set model had to be written. To give you an impression on the size of the task, here some numbers: The original Question2Answer framework consists of about 36’000 lines of code. Until today, I have written 18’100 new lines of code for PhysicsOverflow in form of plugins, layers, overrides and changes in the core code, about half of the size of the system. The new system required 5’800 additional lines of code, which explains, why it took that long to realize it. I hope now that I was able to test all use-cases so that there remain eventually only minor bugs.

The next development step will be a surprise, stay tuned!

polarkernel

PhysicsOverflow is Living and Animated

Wondering what happened on PhysicsOverflow while I have developed the preliminary phase of PhysicsOverflows reviews section, Reviews I, I have looked on the host providers statistic and have done some queries in the event log of our database. Here are some numbers illustrating the pleasant activity on our site:

Number of visits in May 2014 (provider’s statistic):

visits_may_2014

As you can see, the number is slightly growing. The number of visits has been above 300 all the time approaching now 400 by end of the month. Maybe somebody has an explanation for the peak after the second weekend?

In the event log table of the database, 14’482 events have been logged in May 2014. Some interesting numbers are

  • 4126 times, a user has logged in.
  • 200 questions have been written.
  • 353 answers have been posted.
  • 1002 comments have been added.
  • 4995 upvotes have been given.
  • 86 new users have registered on the site.

I think this is really not bad for the second month after publishing the site. Maybe we will already have quite a number of submissions in the reviews section by end of next month? Interested? Submit a paper to PhysicsOverflow!

Admin Dashboard Settings: Part 5 – Posting

The menu Posting of the Admin Dashboard allows to customize different things related to posting questions and answers, as you can guess from the name 😉

Some basic options can be enabled or disabled by a check mark. I write down here what personally would choose, so please say so in the comments if you disagree

  • Close questions with a selected answer: nope, and in addition until now we rather tended to refrain from choosing a selected answer anyway
  • Allow questions to be manually closed: yes
  • Allow users to answer their own question: yes
  • Allow multiple answers per user: yes
  • Allow questions to be related to answers: don’t know what this means
  • Allow comments on questions: yes
  • Allow comments on answers: yes

Then you can choose either the Markdown Editor or a WYSIWYG Editor as the default for posting questions, answers, and comments. Here, I think we should choose an appropriate Markdown Editor, see also the discussion about Plugins.

Next, you have the possibility to provide custom messages that appear in the forms used to ask, answer, and comment which can be activated by a check mark. These features are similar to what Stack Exchange applies to overpatronizing users by writing “don’t say thank +1” there etc … ;-).

  • Custom message on ask form – HTML allowed
  • Custom field for extra information on ask form
  • Custom message on answer form – HTML allowed
  • Custom message on comment form – HTML allowed

Some minimal criteria posts have to fulfill can be specified next:

  • Minimum length of question title: 
  • Maximum length of question title: maximum possible is 800
  • Minimum length of question body:
  • Minimum number of tags: 1
  • Maximum number of tags:
  • Use commas as the only tag separator: yes (check mark)
  • Minimum length of answer:
  • Minimum length of comment:
  • Check email notification box by default:

As this has nothing to do with political moderation ;-), I think we can use the SE settings where possible here.

After this comes a text field for censored words, you have to separate them by commas. As we are, conversely to what Stack Exchange thinks, grown up intelligent people who do not need to be patronized by telling us how we are allowed to communicate with each other etc, we probably don’t need this. However, be writing some awful political terms SE often (ab)uses there, such as “broken window”, “big city problem”, “not constructive” etc, this feature might be useful as a detector for SE politicians who are trying to undermine our site  … ;-P. That the SE typical political overmoderation can indeed completely destroy several years old beta sites by continuously and strongly acting against the community, you can see for example here.

Ok, back to business 😉 … Then come some options related to define related questions:

  • Check for similar questions when asking: yes (can be enabled by a check mark)
  • Similar questions matching: here you can choose widest, wider, default, narrow, narrowest
  • Maximum similar questions to show: maximum 50

Finally, you can specify some settings concerning tagging :

  • Show example tags based on question: yes (even though we do not expect to have as many askers as Physics SE who do not even know enough physics to choose appropriate tags and therefore create tons of silly unprofessional ones 😉 …)
  • Example tags matching: widest, wider, default, narrow, narrowest
  • Show matching tags while typing: yes (can be enabled by a check mark)
  • Maximum tag hints to show: maximum 50

Generally, I think we should seriously discuss what Physics Overflow urgently needs before going (bugfree LaTex, TP users can reclaim their posts and accounts, attribution of questions not taken from the TP data dump, …?) online and which less important but still nice to have technical issues can be deferred a bit and resolved while the site is already running, in the near future, maybe before Christmas.

Admin Dashboard Settings: Part 3 – Permissions (and first thoughts about community moderation in Q2A)

In this post I’d like to explain the privilege system Q2A offers, and then motivate a discussion about what is the best way to set up moderation. At the beginning we will have to deal with the handicap that there is not yet a  large community (or  promotion of the site will have to work exceptionally well) and many or most of us will not yet have the rep corresponding to their knowledge and expertise, which makes community moderation based on rep dependent privileges a bit difficult.

The settings concerning the privilege system of Q2A can be accessed under the menu point Admins/Permissions .

There is the following hierarchy of roles, users of the site can take, I’ll list them starting with the ones which should be the least powerful:

  • Unregistered users ( := anybody)
  • Registered users
  • Registered users with enough points
  • Editors
  • Experts (technically created by Moderators and Admins)
  • Moderators (technically created by Admins)
  • Admins (technically created by the Super Administrator)
  • Super Administrator (created during the process of initializing the database as far as I understand it)

The permissions for the following moderation actions are fixed in Q2A and can not be changed (without the help of a good hacker 😉 …)

  • Blocking or unblocking user or IPs: Moderators and Admins
  • Approving registered users: Moderators and Admins
  • Creating experts: Moderators and Admins
  • Viewing user email addresses: Administrators
  • Deleting users: Administrators
  • Creating editors and moderators: Administrators
  • Creating administrators: Super Administrators

These are some kind of Super-Powers that should really be only accessible to people the community fully trusts, if that privileges should be used unilaterally at all (?) … And of course there is nothing that prevents us from deciding in a community driven procedure who should be the experts, moderators, admins, if and when users should be blocked, etc by discussing, polling, or other means in the appropriate Meta (sub) category.

For other privileges, the permissions can more or less flexibly be customized (the / means or):

  • Viewing question pages: Anybody / Registered users
  • Asking questions: Anybody / Registered users / Registered users with enough points / Experts, Editors, Moderators, Admins
  • Answering questions: Anybody / Registered users / Registered users with enough points / Experts, Editors, Moderators, Admins
  • Adding comments: Anybody / Registered users / Registered users with enough points / Experts, Editors, Moderators, Admins
  • Voting on questions: Registered users / Registered users with enough points
  • Voting on answers: Registered users / Registered users with enough points
  • Voting posts down: Registered users / Registered users with enough points / Experts, Editors, Moderators, Admins
  • Recategorizing any question: Registered users / Registered users with enough points / Experts, Editors, Moderators, Admins / Editors, Moderators, Admins
  • Editing any question: Registered users / Registered users with enough points / Experts, Editors, Moderators, Admins / Editors, Moderators, Admins
  • Editing any answer: Registered users / Registered users with enough points / Experts, Editors, Moderators, Admins / Editors, Moderators, Admins
  • Editing any comment: Registered users / Registered users with enough points / Experts, Editors, Moderators, Admins / Editors, Moderators, Admins /Moderators, Admins
  • Editing posts silently: Experts, Editors, Moderators, Admins / Editors, Moderators, Admins /Moderators, Admins / Admins
  • Closing any question: Registered users / Registered users with enough points / Experts, Editors, Moderators, Admins / Editors, Moderators, Admins /Moderators, Admins
  • Selecting answer for any question: Registered users / Registered users with enough points / Experts, Editors, Moderators, Admins / Editors, Moderators, Admins /Moderators, Admins
  • Viewing IPs of anonymous posts: Anybody/Registered users / Registered users with enough points / Experts, Editors, Moderators, Admins / Editors, Moderators, Admins /Moderators, Admins
  • Viewing who voted or flagged posts: Experts, Editors, Moderators, Admins / Editors, Moderators, Admins /Moderators, Admins / Admins/ Super Admins /
  • Flagging posts: Registered users / Registered users with enough points / Experts, Editors, Moderators, Admins / Editors, Moderators, Admins
  • Approving or rejecting posts: Registered users / Registered users with enough points / Experts, Editors, Moderators, Admins / Editors, Moderators, Admins /Moderators, Admins
  • Hiding or showing any post: Registered users / Registered users with enough points / Experts, Editors, Moderators, Admins / Editors, Moderators, Admins /Moderators, Admins
  • Deleting hidden posts: Editors, Moderators, Admins /Moderators, Admins / Admins
  • Posting on user walls: Registered users / Registered users with enough points
  • Vote on comments: Anybody / Registered users / Registered users with enough points / Experts, Editors, Moderators, Admins / Editors, Moderators, Admins /Moderators, Admins / Admins

As you can see, some privileges can be made rep dependent similar to how it works on Stack Exchange. We already said that accepting answers is probably not needed, and indeed it seems that in Q2A this feature would give people the power to accept answers to not only their own questions (will have to check this). The yellow features and privileges are not known in Stack Exchange.

A possible way to initialize a rep dependent privilege system is what SE does, to make only viewing questions, asking, and answering possible for free, whereas  allowing other things dependent on rep (or points). One has to be careful, because some privileges do not behave in the way known from Stack Exchange. For example as it stands, there seems to be no possibility to roll back edits, I am not sure if this is due to some still open issues with the edit history or if we will have to find another solution for this. Closing and reopening questions is unilateral, such that the action immediately kicks in if users, who have the power to, push the button. Approving or rejecting posts I do not yet fully understand …  Hiding and showing posts seems to be some kind of “soft deletion” which you can do on SE with your own answers for example. But again, in Q2A the people who are given the power to do it can not only “soft delete” their own posts, but any questions and answers. Hidden posts can obviously get “hard deleted” to getting rid of them. I am not sure about the technical difference between hard and soft deletion, what effects the two things have on the underlying database etc (maybe Polarkernel knows …  😉 ?)

As already mentioned, setting up a good fair community moderation  might be a bit a challenge at the beginning, when the community is still small and there are not yet enough highly reputed users who can efficiently make use of a rep dependent privilege system. A possible (potentially stupid 😉 …) workaround that comes to my mind is to make heavy use of the possibility to poll in meta “list questions” ;-P  about higher order moderator actions such as closing / reopening /deleting / etc. In fact, Dimension10 was faster than me LOL 😀 and has just written an article about community moderation by meta posts  here
To this I basically like to add that we can steal the Requests for Reopen Votes thread from MathOverflow (and Maths SE) and make a similar one for Requests for Close Votes, too. The idea is that if you think a question should be reopened you can say so in an answer and shortly explain why. Other people can then vote on this answers and discuss the issue in the comments below. If the score of the answer passes a critical positive number X we agree on, say for example +5, somebody who has the power to reopen can do it and edit the notation [reopened] into the first line of the answer to say that it is done. A request for close votes thread could work in the same way, such that a question mentioned in an answer gets closed, as soon as the score of the answer passes for example +5, too.  The other things suggested in Dimension10’s post ( tag synonyms, “burnination” what silly word is this, it is NOT in the dictionary …? Should int not rather be burnication …? of tags,  FAQ questions, suggesting categories, etc) could work in a similar way. Not sure, if we would need some kind of “Request for Delete Vote” meta thread too …? Obviously bad things such as spam, offensive and insulting stuff, crack, etc should probably be deleted faster (?) by handling the flags, otherwise for non-bad things I think we do not want to get infected by the Stack Exchange deletism 😉 and rather preserve things including comments and “robust” discussions which are allowed (!), as they do on MathOverflow .

Admin Dashboard Settings: Part 2 – Points (rep settings)

As Polarkernel pointed out  here, there are some issues concerning the correct installation of the votes for from TP.SE imported posts (and any posts imported from an SE data dump ..?) .

This has among other things the effect that it is not straight forward to recalculate them after for example a decision to change the rep a user earns for an up vote on his question from 5 to 10 (as it was when I joined Physics.SE …). So these rep settings for different events as up/down votes, accepts, etc should be settled before importing the TP questions. As some people said, things that worked on Physics SE should not be changed, so I suggest to apply mostly the same settings. The points for different events can be customized by going to the Admin Dashboard and chose the menu Points. I will now list the available options, together with the SE settings:

  • Posting a question:   0 points
  • Selecting an answer on your question:  2 points
  • Per up vote on your question: 5 points
  • Per down vote on your questions: -2 points
  • Limit from up votes on each question: ? not available on SE
  • Limit from down votes on each question: ? not available on SE
  • Posting an answer: 0 points
  • Having your answer selected as the best: 15 points
  • Per up vote on your answer: 10 points
  • Per down vote on your answer: -2 points
  • Limit from up votes on each answer: ? not available on SE
  • Limit from down votes on each answer: ? not available on SE
  • Voting up a question: 0 points
  • Voting down a question: 0 points
  • Voting up an answer: 0 points
  • Voting down an answer: -1 points
  • Multiply all points by: 1x
  • Add for all users:  1 point

BTW isn’t it some kind of fun that we can now choose these setting on our own … 😉 ?

Concerning the limits from up/down votes of question and answers, I would personally set them to a large value such that these cutoffs have no impact and our site is renormalizable … ;-P.

General Remark: I think it would be a good idea to write the settings we agree on in these Admin Dashboard discussions down in the respective posts, such that we have them ready when we want to go online. This holds for Dimension10’s post about the General Settings too.

Things that should be done better on the new site than on Physics SE

As discussed in the comments to previous posts, the general Stack Exchange concept of running a Q&A site works well and the features available are mostly useful,  so things should not unnecessarily be changed too much.

However, on Physics SE there are certain things that are handled in a way which is rather counterproductive to building up a free academic community, which should therefore be avoided / done differently on the new site.

Here is a poll to identify what exactly these things are:

This list is most probably not exhaustive (and not too well-organized either) …

So everybody is welcome to point out in the comments what is missing or not formulated accurately enough, such that the list can be completed and improved.