If a user is referencing his site, is it sufficient for him to disclaim his connection to that site in his profile - or should he disclaim it in each post where he is referencing it?
As the point came up concerning the ReplicationWiki I'd like to point out it is not a private website. We founded it at our university, but there are more than a hundred registered users, a number of whom have actively contributed, and it is a non-profit scientific project that just has a lot of content that helps to answer questions like some of those asked here. It seems to me that some users here just don't work with scientific data and this is why they don't see that. The question for instrumental variable datasets for example had two answers. Both can be found in the wiki, now also with a link to the valuable literature that was mentioned here. And the wiki has 100 additional examples. Such information can be extremely useful to researchers. However, after the suggestive way that the question was asked if this is "spam", including a claim that the information is "usually of weak relevance" (in spite of many upvotes for various answers before), two people voted this particular answer down.
I clearly see the point about editing previous answers and that in general one should be cautious about one's own bias but in my eyes the value of contributions should be primarily judged by the content that is added, and if there is something questionable it would be nice to approach users in a respectful way, especially new ones who openly use their real name. Imagine the founder of wikipedia had come here before wikipedia was well known and had shown that for a number of questions here helpful information can be found in that wiki. Would that have been a bad thing?
The general answer to the question I have now found in the guidelines: How to not be a spammer: "Post good, relevant answers, and if some (but not all) happen to be about your product or website, that’s okay. However, you must disclose your affiliation in your answers". As that is the rule here I've now followed it and marked my affiliation in every one of my contributions.
What if I see someone doing something bad? Flagging comments/answers/edits. I guess then the user is informed and can react, and there should be no need to make it public. Directly approaching someone (especially new users) politely would still be nicer but I can also understand that it has its advantages if in an online community users have their privacy and don't need to look for messages, espeically unsolicited ones, also because not everyone uses such communities very frequently.
I think a bold disclaimer is necessary whenever a user talks about a product that he is affiliated with. That is, any reader (OP, or future visitors with the same question in mind) should be always aware of the relationship.
- A disclaimer in the profile is not enough, as we do not require users to check the profile after reading every answer to be aware of potential affiliations
- A subtle disclaimer, i.e. "our XYZ" is also suboptimal as it is not "bold". The reader should be without reason be aware of the connection
- Therefore, concluding, a form of "Disclaimer: I am working for/on XYZ" or "Have a look at XYZ (which I am working on)" is necessarily the conclusion of these two requirements.
And I would say this is necessary whenever the user is advertising a product/project, no matter whether it is for profit or not. In either case, incentives are not aligned: the user may want to present the product better than it is.
Also, when an unrelated third party is referencing XYZ "have a look at XYZ", this information carries more value than when a member of XYZ is referencing it. Both because the third party has less incentives to post it, and also because a third party being aware of it intrinsically conveys information.
Finally, there is never a down-side to such a disclaimer. More (relevant) information is better. If someone's reply is "but I'm having so many posts and comments about XYZ, I can't really be expected to add the disclaimer to each of them", he probably is spamming too much about XYZ.
While I don't think having the disclaimer in rarely viewed profile info is enough, I think the "our XYZ" disclaimer in every answer is a good enough solution. I have seen it used in several answers and I think it was always quite clear.
And even though there is obviously competition in science as well I don't think disclaimer requirements for non-commercial products need to be too strict if the product is raised with good reason.
Specifically with respect to the RW case I think the RW is a very useful tool and Jan Höffler can contribute greatly to Economics SE so I would not knitpick too much over the details.