I just recently switched to the Mac OS from XP (thank God)and I had been using Copernic Agent for my intensive internet searching. I have a copy of CA installed under Parallels desktop and today I did the first of a couple of comparisons I want to do.
I generally do two kinds of intensive searching. The first is where I want to find everything possible on an obscure name or group so I don’t expect many results. I did such a search today with both products under such a scenario and I was happy to see that DA found almost the identical results. On the good side for DA, I didn’t need to edit out worthless pages. However, because DA is actually reading the pages, the search took much longer but that is unfair because I didn’t set CA to “analyze pages” at which it does a lousy job anyway.
The drawback to DA here is:
It doesn find PDF or anything other than HTML. Hopefully that gets fixed soon.
Because it weeds out pages without the exact term, it rejected some pages where the name was in inverted order whereas CA reports the page just as it was found by the search engine. This is unavoidable given the nature of what DA is doing but I will have to remember to be more careful about creating queries to cover this eventuality
Overall, I was happy with the breadth of the DA search. I think DA will really shine under my second scenario which involves searching for everything that can be found where the name or group is not so obscure. In searches like this with CA, I routinely got 2500 results, a huge percentage of which were garbage and CA it is very difficult and time consuming to edit these using the program. I can see that DA will really help in this case, and I will post feedback later.
Overall, this is an impressive program and when support for other data types is added, I think I can pretty much ditch Copernic. It hasn’t been updated in almost four years, so I was getting pretty tired of it anyway. Many of the modules for the various search engines consistently report only garbage when using anything other than simple queries.
Instead of searching for “Harry Potter” use (Harry NEAR/2 Potter).
DA will then look for pages that contain “Harry” within two words of “Potter” and so will pick up both “Harry Potter” and “Potter, Harry”.
DA has very powerful Boolean operators, with provisions to add a second line to the query formulation and to select the drill depth of a search.
Play with the features under the Digest tab of a completed search. DEVONagent’s AI features create a dynamic list of “topics” based on the content of the found pages. Click on a topic and play with the visualizer buttons, which – depending on the content of the pages – can be very useful.
Note also the page summaries provided in the Digest view. You can save the summaries (including hyperlinks) as rich text.
And of course note the interactions between DA and DT Pro. As DA has already downloaded the found pages, it’s an easy matter to transfer all or selected pages to a DT Pro database. Or to initiate a DA search set opening from within DT Pro.
Christian is aware of requests for searching/downloading PDF (and perhaps other) file formats.
I am aware of the NEAR operator but I was trying to do a symetric comparison. Actually, in the process of doing this I went back to Copernic and discovered much to my chagrin that it also has the capability of processing results by the search query. Now I really feel dumb because for years I have been doing that manually. I can’t believe I overlooked that.
That also means that one of the advantages of DA over CA just vanished, so I will have to rethink this comparison at a later date. I promise to take into account the features you mentioned, but I have to say that the ability of CA to process PDF and other data types plus a better of way of updating and tracking old search results are kind of tipping the balance for me at the moment.However, I really like the way that DA lets you view the results in the built in viewer rather than having to open the page in a browser to see the results.
Anyway, I will post another comparison when I have time to really use DA and think about all this. In the mean time, with an Intel Macbook, I get the best of both worlds!
Just wanted to let you now that, coincidentally, I just got a project that will require me to go through a large amount of documents that I had accumulated a couple of years ago. I need to go through the documents again in order to classify them and find some structure. I immediately though of DevonThink and I fooled around for a bit after importing them into a database.
I was quite impressed at first glance after doing an auto group and finding that the program have a a good head start. I look forward to doing more tommorrow and in the coming days.
I’ve started several databases by dumping thousands of documents into them with no initial organization. Depending on the nature of the contents, auto grouping can sometimes get me off to a good start, at least for some of the contents. Moving Search results (or selected portions of the results) into a new group can be a big help, too. it can often be useful to replicate search results into a new group, as DT Pro allows one to locate replicants in two or more groups. At some level of organization, the Classify button starts making useful suggestions.
I’ll confess that for some projects I never fully finish an organizational structure, as the See Also button works reasonably well even with a somewhat sloppy database. And of course searches don’t depend on a detailed organization.
After a couple of days of work with DT, I have to say I am very excited. I have used just about every program in this genre on Windows including Onfolio, NetSnippets and a less known program that is probably the most comparable which is AskSam. None of them come close to DT. In fact, I like it so much that I am prepared to overlook, for now, the fact that DA can’t directly search file types other than html.
I found a work-around which is to do my main searches on DA, importing into DT of course, but then doing a supplementary search in Copernic Agent (CA). CA does have superior filtering and grouping capabilities compared to DA so I can filter out all html results leaving only PDF, DOC, XLS, etc. I can then spit out an html page with the results which I can browse or even clip links for import into DT. Its not perfect, but it will have to do until DA is able to search these file types directly.
I still plan to do a side by side comparison when time permits.