Same weird info here: wsp (Bill) considers RN a real good program, up to the point of buying it (except for some bug he spoke about some time ago, without answering the question if that bug persists), and PIMfan speaks of RN 6 whilst the current version seems to be 3.0.6.

Anyway, I trialled it again today, and I must say it's AWFUL.

1) No clones. Ok, we knew this, it is deemed to have an elaborate tag system instead. Accepted. Problem is, the tag system is not really good, to say the least.

Ok, you can both rename and merge tags, that's very ok. But assigning tags by kb is not possible, you'll have to move your move rather extensively. Then, it might appear unbelievable, but it seems it's impossible to "search" for tag combinations (neither by the tag functionality, nor by the search one, hence the "").

There are special "folder tags", i.e. you assign those to sub-headings, let's say a geographic classification (list of towns), under which then you place your customers, and those "folder tags" will be inherited by the children (here: customers in the respective town), be them present at that time or be them newly created afterwards, and the above renaming/merging functions properly here, too. So all this is very "inviting".

But then, in real use, tags are a total mess (3 different panes in the special F9 pane just for tags), and in 1 hour's "work", with the help of the help file (there is no forum for bauerapps), I wasn't able to see the horizon here, and it doesn't seem possible to just see all your items tagged "ab" AND "bc", let alone better combos, let alone OR: Not even ab AND bc seems to be possible, which means this tagging system is WORTHLESS.

If I just didn't understand how to do it, please tell me.

2) Then for searching, i.e. the same task as above, but with simili-tags like ab, bc, de, whatever within the text (or #ab, $bc or whatever). Here, the help file doesn't meet the real program:

"All terms (for FTS only) - Match all terms together or if unchecked then match any term"

So for all possible Boolean search, RN just seems to have AND and OR, the latter being the default.

Well, I switched forth and back from "free" to "Prof trial" (and closed the program in-between) but did NOT see any option "Match all terms" or anything, the search options in both cases just being

Search mode: normal, db, fast
Search in: captions, content
Options: match whole word, match case, exact phrase
Scope: current page, all pages, include protected

That's all (and there's an additional button to search, in Prof. version, within all open files/trees).

Then, of course, I try manual searchterm1 AND searchterm2 (and the same with OR), with NO result, since just a searchterm1 searchterm2 will not bring any result either:

Let's analyze this further: st1 st2 SHOULD bring results whenever there are items containing either st1 OR st2 since that is deemed default, but in reality, no result even here (which means that "exact phrase" seems to be ON anytime, even when it is NOT checked; no result either when there are even items containing BOTH words, but not as a phrase, and this again indicates there is a BIG BUG which makes the "exact phrase" working anytime

I tried this with regular ascii chars, i.e. ab, cd, and such, not even öä, éà or whatever might cause special probs.

So, considering RN is deemed to work on SQLite, which is a very professional albeit tiny db, its search functionality must be declared ABYSMAL.

Or then, I did not understand that the "FTS" = "full text search" has to be triggered in some special way which I didn't find, so please tell me, but from all I read in the help file, FTS is ON whenever I check those options and then click on the Search button.

3) This is negligeable in light of the above: Many functions ain't accessible but by mouse.

3a) Ditto: Even in the free version (which seems very "generous", i.e. not that much crippled, but that's without interest if even the prof version doesn't work properly), you can put links to external files into the tree, but there doesn't seem to be any way to trigger that link but by mouse, again. But then again, in light of 1) and 2) this program seems to be really worthless, in spite of such little goodies as here in 3a (even without kb triggering), or hoisting, or whatever: The core functionality to be expected here doesn't work properly.

Or then, please tell me how to do it right.

Just in order to clarify: "forth and back" is not possible, the unique installer installs the free version, and once you try the prof version, going back to the free one isn't possible anymore... except perhaps automatically, after 30 days, but not even that is for sure since the web page says that you must buy the prof version after 30 days. But then, I suppose it will revert to free after 30 days if you don't buy... but I'm very tempted to kick it out this very first day, so I presumably will never know.

What piece of crap.

You've used much more inflammatory language than I would, and I confess I don't have the time or patience to go through each of your complaints. But I would basically agree with you that RightNote is not terribly satisfactory at the moment. I did buy it earlier, but my two main note-takers are still MyInfo and Evernote.

I have found one excellent use for RightNote, however, that justifies the money I paid for it. I attach PDF files (of which I seem to have a lot nowadays) to RightNote, and it then becomes easy to search the contents of a number of PDFs simultaneously. Once I have linked PDFs to a RightNote "page" (i.e. a subsection of a RN file), RN seems to absorb the PDF's underlying text (but not the PDF itself) into its own file, so that any searches thereafter show a list of occurrences of a word or phrase in all linked PDFs, in each case with the word or phrase highlighted and in context. I find that extremely useful.

So I've left the RightNote icon on my taskbar for that one specialized purpose, and of course I keep an eye on any possible improvements to the program in general. But so far it hasn't persuaded me to abandon my other note-taking programs.

Bill

re: " PIMfan speaks of RN 6 whilst the current version seems to be 3.0.6."

Typo on my part. I was looking at RN 3.06 and also at MyInfo 6 at that time and transposed version numbers....

1)

PIMfan, thank you for clarifying.

Bill, thank you for reminding me. It seems the bug you encountered with RN was precisely with its pdf functionality, so it seems to have been sliced out?

What you describe for pdf in RN - if I get well what you say: You just link TO pdf's BUT in the corresponding link item in RN, the TEXT of the pdf is replicated, AND the respective terms are included are into RN'n index. Ultra Recall functions this way, by option, for some external formats, too (and perhaps for pdf's also), but this again is another subject, i.e. HOW external formats are integrated in different programs (I tried to have us clarify this some time ago, but with no real success):

a) external file is imported (db is blown up, nobody really wants this)
- several variants similar to those below

b) external file is just linked to, but cannot be searched from within the PIM

c) external file is linked to, and is made searchable from within the PIM, by way of just including all its terms/words into the PIM's index - we might have a synch prob here (see d))

d) external file is linked to, and its text-only is imported into the link item of the PIM (and hence it profits from any indexing the PIM will have got for internal things, too) - again, we might have a synch prob, i.e. will the PIM monitor whenever you open the external document (aa) from within the PIM, bb) from within its original applic), reenter its text, and thus rebuild the index? Here in d) aa) that might be simple, but bb) is less simple, and the similar problem in the c) case will probably be left out by most developers anyway. Of interest here: The developer could make his PIM CHECK the file data of all files linked to from within the PIM, whenever the user triggers a special command for this ("update links" or such).

Of course the synch problem is not there for read-only pdf's.

2)

My point with RN is

- A db-based PIM must make profit the user from the db layout, i.e. we accept in text-based db's that we don't have clones, tags and other goodies very difficult to realize in such formats, but for db-based PIM, all this becomes easily available to the developer, so he should make it available to the user. Hence my asking more from RN than from some simple TreeNotes 3 (or from Jot Plus).

- The same is true for more elaborate searching, since Boolean search is much simpler to realize with an index, and the index, in turn, with a db. This also applies to the existence or not of a "hit table", and indeed, RN HAS got a hit table, similar to MyInfo, but such a hit table is not that useful if you cannot combine several terms from one item within your search. To clarify: I could perfectly understand that the developer takes such a feature out of the free version, but even the prof. one seems to be both crippeled and buggy, Even (non-db) Jot Plus' search functionality is way better than is RN's.

- And the same is true for tagging: Here again, it's the db concept that makes this "possible" and easy. So RN's got tagging, and multiple tags, of course, but then, you can't gather several items into a group that combines several such tags: It seems to be impossible to gather, e.g., several "big" customers (with size tag 3 out of 1, 2, 3), from several towns (in the above example), let alone have a third or fourth tag, e.g. such customers but in a certain industry only, and which you should contact again this month (presuming you have some 6 or 12 different "contact again" tags for the months to come).

- And as said above, this is not possible with tags, and it's not possible with search either since the search doesn't allow for AND (and not even for OR - this MUST be a bug!)

So RN does not offer any "db advantage" to the user, except for the "hit table"... from which then you'll manually pick those items that meet your additional search terms that RN doesn't let you enter into its db search.

That's devoid of sense. Btw, about 2 years ago, I had trialled RN, and now I remember very well that even then, it already had its weird 3-panes-for-tags-alone, but no possibility to "search" for a tag combi.

In short, they do a db-based PIM, and so you think you'll get a db-based PIM, i.e. one offering the usual functionality going with such a thing, and then you won't get that, or only in minor parts: indexing of external docs is there, ok.

But Bill, even for your external pdf M:
- The above-mentioned search bugs should apply!
- Not speaking of tagging external pdf's and then wanting to search for combos of those, which could come very handy but isn't possible with RN.
- And isn't it awful to have to click on the view button, in the content frame, for viewing every single pdf, instead of there being a shortkey which'd allow for just getting to it from browsing within the tree? (But no control/shift/alt-Enter or such will do.)

This discussion is way over my head, as usual, but I'll try to respond briefly. As I understand it, RightNote uses method B in your list. In other words, it imports and indexes the underlying plain text, but it does not import the PDF itself. This seems to me a better way of dealing with the problem than Evernote's trick up swallowing up the entire PDF, because that means your Evernote file can become enormous in a hurry if you're importing many PDFs. RightNote will, if you ask it, also show the plain text of an entire PDF. That's handy once in a while.

I agree that RightNote seems rather clunky at times, and I wish I could find another way to do quick searches in groups of PDFs. Perhaps you or others here will have some suggestions. Of course I'm aware that I can use the search box in Windows Explorer, but that only produces a list of files, not snippets of actual text. I need the latter.

Bill

WSP wrote:

I agree that RightNote seems rather clunky at times, and I wish I could
find another way to do quick searches in groups of PDFs. Perhaps you or
others here will have some suggestions. Of course I'm aware that I can
use the search box in Windows Explorer, but that only produces a list of
files, not snippets of actual text. I need the latter.

You could put the groups of PDFs into separate folders and then use PDF-Xchange Viewer to just search the PDFs in that particular folder. PDF-Xchange Editor can do the same, with some extra search functions. Same could be also done with Agent Ransack, with the added benefit being that the contents of the entire paragraph are shown where the search term occurs.

Btw, Agent Ransack has had an update recently, first time in 3 yrs:
http://www.mythicsoft.com/agentransack/history

While I too love and prefer PDF-Xchange Viewer, it's worth mentioning that you can also use the standard Acrobat reader to search through a folder plus all its subfolders.

Interesting to hear about that RightNote feature, however: I don't rate the program very highly, but that rather clever simple-text function sounds quite cool. As long as it supports accented characters, that is (all too frequently a weakness of outliners and the major reason I don't use Surfulater).

Cheers,
Bill

Given the choice you pose, I would say that RightNote is not total rubbish.

"Am I just dumb, or is RightNote total rubbish?" - Well, if you ask this way, my answer is: Yes.

Until now, I have only used RN for importing notes from Evernote (see the thread I opened a few weeks ago). As the project, I need RN for, is delayed, I haven't dug deeper into RN. Importing worked (apart from a glitch that I could easily "repair" - RN does not import notebooks with umlauts in the name) - so for me RN is no rubbish at all.

So, a more precise answer to your question is: I don't know (re dumb), and: No (re rubbish).

Franz

Dr. Andus, thanks for the suggestions, but I use PDF-Xchange daily and have tried Agent Ransack in the past. Both of them do the job more or less, but they are rather slow because of the lack of indexing. Despite our grumbling about RightNote, searching through a big stack of PDFs is one thing it does extremely well -- and even elegantly.

Bill

WSP wrote:

less, but they are rather slow because of the lack of indexing.

Have you tried Copernic Desktop? It does do indexing, the search is quicker, and there are some options to search in specific folders, file types etc., and you'll see the search term in context.

Thanks again. I'll look into it.

Bill

1)

Bill (wsp),

I even had been tempted to give that advice given here, without being asked, then withheld it, since I don't know X1 (x1.com) well enough, and it's not a real elegant solution. Here, Copernic Desktop (Cop) has been mentioned, instead of X1; Lookeen (lookeen.net) should be a third solution (and perhaps the best one (?) if you also use OL (Outlook): Lookeen was THE OL "specialist", and is deemed to provide the best search in OL, whilst Cop is deemed so-so, and X1 seems to have trouble with OL (but if you do NOT use Outlook, X1 could be your best choice within this context here since for anything besides OL, it seems to be better than Cop, from what I've read on the web).

Do NOT overlook Lookeen (in case you use OL) since now it's a more-or-less full-grown desktop search engine, i.e. also for Word, Excel, etc, and pdf.

Since you've got all your pdf's there, you should trial all 3 programs (and perhaps more, but dtsearch is so expensive... ;-) )... and expose your experience with each... ;-)

From X1's marketing: "X1 Professional Client is a desktop search application that is specifically designed to work with Outlook." - this leaves me doubtful, since people report probs with OL, and in fact, that was the reason I did not (yet) buy X1... (using OL myself)

Btw, there doesn't seem to be very active development neither with Cop nor with X1, so compatibility probs could remain, instead of being overcome...

2)

Bill, if I understand your problem well, it's caused by MI's inability to do what you now do with RN; both UR and TB do it (attention, only UR Prof might do that, and it's only some weeks ago they had their annual bits offering), though, but I'll not try to make you switch to either of them; none of them is that good, in other respects, so you'd not switch anyway, and you'd be right to stay with MI, especially since Petko's new data format might change this non-processing of pdf's.

Or in other words: Today, he would be crazy to buy some add-in for that function; he's got better things to do; but if he doesn't buy such a pdf component after having changed his db format, well, that would be inexcusable since that new format (SQLight? does he say anything about WHAT it will be?) will be MI's format for a very long time, so a minimum of investment (such a component should cost him about 500 bucks) should be possible.

3)

Bill, "I use PDF-Xchange daily and have tried Agent Ransack in the past. Both of them do the job more or less, but they are rather slow because of the lack of indexing". AG/File Locator Light/or even the paid version doesn't do indexing, and for a bunch of pdf's, that's not beautiful. But there are SOME pdf tools that DO indexing, e.g. the one from Nuance (which does as much bona fide customer checking as does Directory Opus, some people call it a nightmare); I even looked them all up but none of them is below 100 bucks or so, it's regularly the "superior" versions. I could look again into it if you want to trial them (some weeks ago somebody here had promised to share his experience with Chaos Intellect without doing so then, but that definitely wasn't you! ;-) )

From all of what I now, Acrobat is awful for your purpose. Hence, again, my "advice" (= that's what I'd do) to trial all 3 progs mentioned in 2) above, all the more so since I do NOT see any real advantage of those "indexing pdf progs" here in 3) over X1 (if that works well with pdf's) since every such pdf prog is mainly for editing purposes and such, and you simply need perfect search, then display, and I didn't find any pdf prog specially made (and optimized) for that task.

4)

Bill, some Boolean (!) search within pdf (UR, TB, and X1, etc. should all offer that, whilst RN does not, because of its apparent bug) is one task; might "clones" of those pdf files within several sub-folders be another, additional task, being worthwile? Or some tagging? Some tagging by coding in the file names? That's two questions, and the answer to the second one might depend on the answer to the first one: IF you've got several sub-folders, THEN both cloning and "tagging" / file name coding could be a good thing; if that was not your choice, you will certainly not judge it worthwile to add tagging/name codings now.

IF some of this could be of interest, let's discuss this in a new thread: There is "Search Everything", or some file managers which are better than others... (And "cloning") of files into different sub-folders could be done by a 1-key ahk command.)

5)

Franz, EN import is the ONLY useful import that RN offers; if you want to import anything else, from other outliners, possibilities are so few that you'll end up writing a script that does it by the clipboard.

My question was indeed a rhetorical one, but in that sense I wasn't entirely sure if RN really was so "dumb", or if it was I who didn't grasp HOW to do it, for those missing elements there.

Since you use RN, please have a look into it, perhaps my assertions are partly wrong, and then I'd be happy to hear about how to find that missing functionality.

But if I'm right, it would be high noon for the developer to see into it.

6)

From my discussing multiple tagging above, and then the need to filter by multiple tags again, here in the RN context, it occurred to my that yes, I had been perfectly right when discussing tags vs. outlines some weeks ago here, and some months ago here, but for the only (and very important) context I had had in mind there: Whenever you have "material", "reference material", "work" and such, then, yes, and I "proved" it there, (rather flat) outlining is highly superior to tagging.

But here above I discussed tagging for "db's" with highly standardized content, my example was a customer db. And here, in such "uniform content" data bodies, a tagging system is as good as an outlined system, or even better.

I think it's very pleasant to have an outline for your very first criterion, but I say "pleasant", not "useful", because I acknowledge that within such "uniform db", it's quite arbitrary which criterion will be your very first one, replicated within the tree: In the above example, the tree would have been by geography, but it could have been by customer "size" / "potential" / "industry", and many more, so a tag-only system (or an outright classic db) would be as good it seems.

All the more so since then, for your second criterion down, you'll need either clones=virtual parents or tagging anyway.

Let's clarify one misconception here, though: Cloning = assigning an element to a virtual parent / stepparent is NOT necessarily more cumbersome than tagging, it's just the execution of the cloning functionality that is often rather cumbersome (but which could be straightened out by an external macro).

BUT then, you assign one customer to perhaps 8 such virtual parents, from perhaps 60, and so you'll have 60 such sub-headings in your tree, every one with its respective virtual children, but NONE with any criteria combination, and THAT would be of interest in real life, most of the time, so the practical interest of such multiple cloning of standard data items is highly doubtful, i.e. must be considered inexistent.

Hence the "advantage" of a tagging system for such tasks: It's simply much more elegant (and here again, assigning standard tags must be a "1-click" (or 2-click, in reality, but neither 5-clicks nor mouse moves) thing):

Most of the time, you'll search for tag combis, and within a tree system, that means searching (or the special search for tag combis you can do in MI, its best feature, and I said this years ago; the only problem, to GET there, since regular filtering in MI will NOT filter for combinations - but the functionality is there, albeit highly cumbersome), and you will almost never open and check those virtual parents.

So, tags / standard db's are preferable indeed, for standardized information, whilst the outliner concept is the superior one for heteroclite data.

(Some could say, that message is not new, but it seems worthwile to have dissected why outliners are superior in many cases, even when finding out that they are not superior in all cases. So here again, it's looking after the right tool for the right task, but most outliner-vs.-tagging discussion is about tagging ostensibly superior for both kinds of tags, which is wrong though, and I could even add, taggin is FAR inferior for heteroclite data, whilst outlining is just a little bit inferior for unified data, but that might come as aggressive. ;-) )

"I could even add, taggin is FAR inferior for heteroclite data, whilst outlining is just a little bit inferior for unified data, but that might come as aggressive."

All the more so since for unified data, outlining is not self-contained, but then, down from its very second "sorting" criterion, must rely either on cloning (not elegant) or on its adversary, tagging. So the "just a little bit inferior" supposes the existence of cloning or tagging within the outliner, whilst tagging "does it all", in this context.

So, the outliner concept alone, for the "unified" task, is not even viable. But I think it's rather instructive to see (cf. the two threads on that subject) how inferior tagging is for heteroclite data (and that many people do not even see this easily), and that on the other hand, outlining needs additional concepts for even starting to become functional for unified data.

Most people, most of the time (and every discussion on that subject amply shows this), don't make that clear distinction (neither did I), so it's a good thing to have investigated both tasks in the end.

I called this "standardized" data, "unified data", but there's the very important aspect of "EXCLUSIVE DATA", which is the core aspect here.

This means, you need the dataset for one customer, OR for some other customer, whilst in the "heteroclite" data / "material" / whatever you call it, it's MULTIPLE COMBINATIONS.

That's the criterion by which you choose tree (and then clones) vs. tagging/db.

Details of the superiority of the tree concept for the TO-BE-COMBINED data here: "Surfulater PrevGen on bits, again" =

http://www.outlinersoftware.com/topics/viewt/5198

I hadn't been aware of X1 having brought out a new version in 2013, so perhaps they even resolved their OL probs, and that would very probably make it the very first choice for pdf indexing, from my totally guessing pov here.

Thanks to PIMfan, I reconsidered RN.

http://www.donationcoder.com/forum/index.php?topic=37935.new#new