The requirements for the documentation of laboratory work are roughly defined by the FDA in CFR21 part 11 (link). In brief: You need to prove when you did the work that led to a discovery and you need a signature attached to it. In paper laboratory notebook adding signatures was a no-brainer but in the digital work this is more complex. Another point is that the document should be protected from alterations. Uploading your document to an external repository would already be a huge improvements over keeping a local copy but you would still lack a signature. Check out our paper about our ELN OpenLabNotes (see link), where I discuss these issues in a bit more detail. 

https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfcfr/CFRSearch.cfm?CFRPart=11

http://journal.imbio.de/article.php?aid=274

If you are the sole individual creating laboratory notes, then possibly yes uploading theme to a server seems easiest to do. But if you are creating a Laboratory Information Management System (LIMS) for a research laboratory that does research into areas where the FDA is or will be involved, it would be best to investigate some form of information management system. This system allows documentation of experiments, inventory, sample number, chemical and hazardous waste, animal numbers (USDA) and other parameters that the FDA will want to see if your laboratory will be inspected by FDA or even by university environmental health and safety. Additionally, this information might be needed IRB, and other oversight entities within the university. LinkedIn has a LIMS group where you can find information to begin choosing what type of system will work in your laboratory.

I'll get to a more specific answer to your question in a sentence.  If you want to understand ELNs, LIMS, etc. I've written a book "Computerized Systems in the Modern Laboratory: A Practical Guide" that will give you a thorough discussion of the subject.

The first step in making a choice is a definition of what you need, and your statement seems to cover that.  Assuming that it is complete, an ELN is just a diary / journal of your work, and in your case needs to be able to meet legal IP support requirements.  The product could be as simple as a word processor, MacJournal, or what you are currently using IF it completely meets your needs. You haven't said what your research is, so there are possible open questions about regulatory support, requirement to meet grants, ensure data quality, etc., noted in the previous responses.

Another consideration is the future.  Are there features / needs that your current software doesn't support: indexing, sharing, searching, security, organization, and so on.  Please note, as was pointed out in Grant Peters comment, that ELNs and LIMS are different things, designed fo to meet very different needs.  The names of the products don't differentiate the difference in application and use well enough.

Will you eventually need to connect your software to non-human, external input (instruments, remotes sensors, access via smartphones or remote computers)?  Do you have to coordinate your work with other groups, and if so what does that require?

So there are simple solutions to what you've stated, but you have to be sure of what your needs are now and in the future.  You don't what to implement something and find out in six months that you needs something else and have to import all your notes and data.  

The key first step is a well developed set of needs & requirements, including legal issues.  Everything else builds on that.  One last point: when choosing a solution, consider the needs for systems support - if you need technical help, where will your get it from?

EverNote: http://evernote.com/

Trello: http://trello.com/

> I don't see the advantage of adopting a new software platform when I & my lab members are happy with the ubiquitous software we are accustomed to. If all we want to do is keep a record of when various advances took place, why can't my lab members simply upload their current lab notebook to a repository every week? Would this type of evidence be less robust in a legal sense?

Weirdly, I don't think anyone really answered the question, instead suggesting ELN options. It's been 2 years since your question, and I would be curious to hear what you ended up doing.

For IP purposes, your idea seems just fine. Starting with what you already use and like, and adding an additional layer for IP. Have you looked into timestamping? There are paid and free services out there, depending on the level of complexity you want (e.g. if you want the service to keep the original documents too). Many of those services are compliant with standards in that area.

My next advice would be to contact whoever is in charge of IP at your institution, and figure out a solution with them. It will probably take some patience from your end, as they are likely to just recommend what they already know, and dismiss things you'd like to do. Lawyers don't like new things...

Thanks for everyone who provided feedback!

Charles, my lab is currently using Benchling to maintain digital notes, and it has worked well for documenting how and when experiments were done. (I also like that it allows us to share data & protocols with ease, but that's a different issue.) We print to PDF periodically because there's nothing inside Benchling that would stop the user from going back and altering old entries. We have not invested in the third-party digital stamping/signing process, although I am aware that this is the gold standard for recording progress. My impression is that the shift from first-to-invent to first-to-file will ultimately downgrade the importance of the lab notebook in the legal arena. But I also have a lingering fear that this presumption could bite me in the ass :)

> My impression is that the shift from first-to-invent to first-to-file will ultimately downgrade the importance of the lab notebook in the legal arena.

Yes, I did not want to bring that up because things are not that simple, and there are other reasons why one would want some kind of timestamping (e.g. trust issues). But for patents, it has probably become much less critical indeed.