thanks for trying Mendeley – we also appreciate your criticism! It’s true that there are still bugs and limitations, but we’ll be releasing a major update in the next few days.
May I ask: Did you enter all your publications manually directly on your profile on Mendeley Web? Or did you import your publications into Mendeley Desktop (which should auto-extract much of the metadata), then upload them to your profile page by dragging & dropping them to the document group “My Publications”?
The latter way should be much easier and faster. I acknowledge that the quality of the auto-extraction also needs improvement, and that’s something we’ll be focusing on after the upcoming beta 0.6.0 release!
@victor I did the latter, because I wanted to upload the PDF. I am not sure I can do that from the web site. Yet, it has been terribly slow going. Your metadata extractor did not get anything right. Please download my PDFs (from the site) and try it out. Beside, you will never be able to extract the metadata to my taste just from the PDF. I think not.
So, I load up the PDF, then it gets the metadata wrong. I have then to click on it and manually correct the reference. That is quite a pain! Even now, the metadata is likely mostly wrong. I simply do not have the patience of entering it, checking the result for correctness and so on. It is not fun! I already do it for my funding proposals, in my c.v., in my activity reports…
Having to enter metadata is boring. Extracting it from PDFs alone is hopelessly difficult. You will end up collecting a lot of metadata, but mostly wrong metadata. No value there.
In this day and age, with Google Scholar, with arxiv.org, your focus on getting us to patiently enter the page numbers, the volume numbers… is misguided. Let me point you to arxiv.org and retrieve automagically my pub. list. That’s quite easy. You can also eat the data from DBLP. I do not think that the value of your tool can be in the pub. lists.
I absolutely agree – the purpose can’t be re-entering all the data manually (that’s the hell we’re all trying to escape)! So I’m sorry to hear that it didn’t work at all with your publications. It seems to be working better for some journal formats than for others – e.g. our users in the BioSciences have reported better extraction quality than users in Physics and CS.
We’re quite confident, however, that we can get it to a good level for all papers soon – both by improving the internal logic, and by querying external databases (such as arXiv).
Similarly, we’ll also be improving the online collaboration and networking features. So the social networking value should increase as well!
I do not tend to reuse any given paper more than a few days a year. So, it is cheaper for me to just find it again with Google Scholar than to carefully achive my papers, or so I think. Once I have the URI of the paper, if I need to write notes about it, I use a wiki with the URI as the paper identifier. As far as maintaining a set of references, I use bibtex (just a simple text file).
By the way, Daniel, what is your opinion about mekentosj Papers?
Thanks,
Mikhail.
Alexander Mikhalevsays:
Daniel, do you think bibtex integration will help you for collaborative work on research project?
I am trying to pull together useful platform for reseachers and would appreciate if you can comment on this
I share bibtex files in subversion repositories. I have been doing it for years. Well, before subversion, I used CVS.
Alexander Mikhalevsays:
Daniel,
given that you can set up social network in a minute by using google friend connect or facebook connect, what value would you like to see from social networking for scientists and what is the advantage of using Mendeley in this case instead of building your own?
Hi Daniel,
thanks for trying Mendeley – we also appreciate your criticism! It’s true that there are still bugs and limitations, but we’ll be releasing a major update in the next few days.
May I ask: Did you enter all your publications manually directly on your profile on Mendeley Web? Or did you import your publications into Mendeley Desktop (which should auto-extract much of the metadata), then upload them to your profile page by dragging & dropping them to the document group “My Publications”?
The latter way should be much easier and faster. I acknowledge that the quality of the auto-extraction also needs improvement, and that’s something we’ll be focusing on after the upcoming beta 0.6.0 release!
Best wishes,
Victor
Thanks Misha, you’re right! I’ve added it to our feature request list.
@victor I did the latter, because I wanted to upload the PDF. I am not sure I can do that from the web site. Yet, it has been terribly slow going. Your metadata extractor did not get anything right. Please download my PDFs (from the site) and try it out. Beside, you will never be able to extract the metadata to my taste just from the PDF. I think not.
So, I load up the PDF, then it gets the metadata wrong. I have then to click on it and manually correct the reference. That is quite a pain! Even now, the metadata is likely mostly wrong. I simply do not have the patience of entering it, checking the result for correctness and so on. It is not fun! I already do it for my funding proposals, in my c.v., in my activity reports…
Having to enter metadata is boring. Extracting it from PDFs alone is hopelessly difficult. You will end up collecting a lot of metadata, but mostly wrong metadata. No value there.
In this day and age, with Google Scholar, with arxiv.org, your focus on getting us to patiently enter the page numbers, the volume numbers… is misguided. Let me point you to arxiv.org and retrieve automagically my pub. list. That’s quite easy. You can also eat the data from DBLP. I do not think that the value of your tool can be in the pub. lists.
The value has to be in the social networking.
Hi Daniel,
I absolutely agree – the purpose can’t be re-entering all the data manually (that’s the hell we’re all trying to escape)! So I’m sorry to hear that it didn’t work at all with your publications. It seems to be working better for some journal formats than for others – e.g. our users in the BioSciences have reported better extraction quality than users in Physics and CS.
We’re quite confident, however, that we can get it to a good level for all papers soon – both by improving the internal logic, and by querying external databases (such as arXiv).
Similarly, we’ll also be improving the online collaboration and networking features. So the social networking value should increase as well!
Probably, a good thing to improve networking will be to add a button “Invite to Mendeley”, by analogy with Facebook.
@Mikhail
I know of Papers, but never tried it.
I do not tend to reuse any given paper more than a few days a year. So, it is cheaper for me to just find it again with Google Scholar than to carefully achive my papers, or so I think. Once I have the URI of the paper, if I need to write notes about it, I use a wiki with the URI as the paper identifier. As far as maintaining a set of references, I use bibtex (just a simple text file).
By the way, Daniel, what is your opinion about mekentosj Papers?
Thanks,
Mikhail.
Daniel, do you think bibtex integration will help you for collaborative work on research project?
I am trying to pull together useful platform for reseachers and would appreciate if you can comment on this
@Mikhalev
I share bibtex files in subversion repositories. I have been doing it for years. Well, before subversion, I used CVS.
Daniel,
given that you can set up social network in a minute by using google friend connect or facebook connect, what value would you like to see from social networking for scientists and what is the advantage of using Mendeley in this case instead of building your own?