Submitted by DragonGem on 2018/10/25 11:39

 Hey Pierre... Now that you've been bestowed Hero Status, how about moving up to DemiGod?

I've been dissecting Paul's excellent Comparison of Note-Taking article by capturing it from WordPress, pasting as Text and assigning Headers 1 thru 4 to mark the various Sections. What I want is to see the whole article BUT with ONLY the IQ-specific sections. This requires a whole lot of Cutting & Pasting to create a hierarchical Grid with each Section in its own Item. Some quick Tagging, Filtering and Exporting and I can get what I want plus the individual Sections can be searched and referenced individually.

It seems to be me that it should be possible to create these Grid Items automatically by simply searching for the Header attributes (down to a specified level) and promoting / demoting based on the previous Section Level. The [Item] text can just be the text of the Header (maybe with the previous upper-level Header appended on). 
Piece of cake

There are a few other packages like Scrivener that allow you to scroll thru a Document and split it manually and I've found a few Word VBA macros that attempt to split Word documents based on Formatting / Section breaks. I've found nothing that does this automatically for HTML files yet. Adding a feature like this on top of the new hierarchical Tags would set IQ even farther apart from other software. Just planting the seed...

I'm going to try quickly building an HTML parser that will break simple HTML files as described above so they can easily imported into IQ. I'll keep you posted on how that goes.

Comments

Hi DragonGem,
 
Yeah, import / export / splitting using styles would be nice indeed...
 
With the current version I did it, at least if I understand correctly the end result of what you want:
  1. Select the whole article
  2. Paste in Notepad and save to file
  3. In IQ: File > Import > non-hierarchical. Open file
  4. Set delimiter to Tab and Import into a blank grid
  5. In this grid, add the Order column
  6. Select the column and do Edit > Renumber (original order of paragraphs)
  7. Sort by Item (right-click header > sort asc) and select all blank lines. Erase blank lines
  8. Sort by the Order column
  9. Column filter on InfoQube
  10. Select all items and set bold font
  11. Remove column filter
  12. Scan the article and select items under each InfoQube bolded entry (there are 7)
  13. Press Tab or CTRL + right arrow to indent
  14. Re-apply the InfoQube filter
  15. Ensure that the column filter setting "Filter out sub-items" is not checked
    (alternatively, you can select the 7 infoqube entries and do a hoist)
  16. Select all filtered items and assign the IQPJM tag (create if necessary)
  17. Remove filter
  18. Column filter for *InfoQube*
  19. Select all filtered items and assign the IQPJM tag (if already checked, uncheck and re-check)
  20. Double-click on the IQPJM Tag
  21. In the Scratch grid, select all items
  22. File > Print Preview > Outline in Bullet List
  23. Click Save and Preview
Results below. I did just a bit more, including section headers and a bit of indenting
 
Pierre_Admin
IQ Designer
 

Pierre_Admin

2018/10/25 12:33

Permalink

  • The four contenders (in alphabetical order) :-
    • ConnectedText
    • InfoQube
    • MyInfo
    • Right Note
  • This review will not give scores out of ten but just compare the programs to each other on the following criteria which I think are relevant :-
    • Writing
      • The comfort of the writing environment
      • The presentation of the text for reading purposes
    • Retrieval
      • Search
      • Favourites
      • Navigation
      • Tagging
    • Big Data
      • Database or File
    • Transclusion & Linking
    • Screen Presentation
    • Ease of Use
  • The comfort of the Writing Environment
  • InfoQube
    • The InfoQube editor is very comfortable to work with. It is a WYSIWYG editor with styles available in a drop down box, at least it is on my system, this program is very configurable and so you can compose your own toolbars with just the commands you need. The ability to select from a number of .CSS files allows you to change the look of the document as you wish and the choice is remembered for each individual document.
    • The editor can be in a floating window which can be placed on a second screen and can occupy the whole of that screen with only a small amount of screen taken up by other things (just the toolbars on the top and left hand side), this is fairly close to being distraction free.
    • The document pane (the editor) can contain various different formats of document but the default is a HTML document whose format is set by a .CSS file. You can have a number of .CSS files for different purposes each with different fonts, layout and colour schemes. I tend to use a very plain one for composition and switch to something fancy once finished.
    • The implementation of tables in InfoQube is adequate but you cannot define the background colours of individual cells without delving into the HTML source code of the page. Borders of cells can be dragged but the results are sometimes not what you expect because InfoQube ‘intelligently’ re-sizes the other cells to accommodate your changes and sometimes ‘intelligent’ can be quite dumb.
  • MyInfo has a WYSIWYG editor with styles selected from a drop down list. Despite this I don’t think the writing environment is as good as InfoQube. The editor feels cramped by all the elements around it, the properties panel can be dismissed but the tree panel cannot. You can open the content of the document in a floating window but this is not editable.
  • The presentation of the text
  • InfoQube
    • In InfoQube you can open multiple document panes in floating windows. By default they are editable which means that you can have more than one instance of the same document open for editing. The question then arises, what happens if you make different edits in different instances ? The answer is one of them will be saved the other one lost.
    • The command to open a new document pane is buried in a sub menu of the ‘View’ menu of the main program which is not as useful as it could be. So I put the command on a toolbar and now it is more accessible.
    • Also the command to lock a pane to a particular document is in the ‘View’ menu of the document pane (there are two sets of menus and two sets of toolbars each of which must be configured separately). It is called ‘Lock Item’ which doesn’t really describe it’s function very well. In my opinion it should have been called ‘Lock Pane’. Anyway it can be placed as an icon on the document pane toolbars.
    • A document pane locked like this is not locked for editing it is just that the pane is locked to showing one particular document.
    • One really neat feature is that if you have many floating document panes open and lock all but one of them then that one becomes the default viewer, if you click on a new item then it is displayed in that pane. If you have more than one unlocked then InfoQube cycles through each unlocked pane in turn as you click on new items.
    • Once you have the configuration of the toolbars sorted out the setup becomes quite useful. You can conveniently view multiple documents in multiple floating panes and refer to one document whilst viewing another. The floating panes can be configured to take up the whole of a screen for convenience of reading or tiled for access to many different texts.
    • I do think that if multiple instances of the same document are opened then the first one should be opened for editing and subsequent instances should be opened as ‘read only’, but that’s just my opinion.
    • Overall InfoQube is an excellent reading environment.
  • Retrieval
  • InfoQube
    • InfoQube has good search facilities. There is an ‘Omnibox’ which searches for a text string in the text contained in the Title or in the Document pane of an item.
    • There is also a ‘Live Search’ pane which does much more, and an ‘Advanced Find’ dialog which can search for a text string in arbitrary fields.
    • There is a Favourites list in InfoQube but it is just a flat list with no separators or grouping. However you can make a grid and call it ‘Favourites’ (or whatever) and set the ‘grid source’ (more on this later) to ‘Favorites’ and the entire list of Favourites appears in the grid and you can then arrange the entries into a hierarchy and as it is a normal grid it is amenable to all of InfoQube’s tools for managing items in grids.
    • The arrangement of documents in InfoQube is not like other note taking programs. An InfoQube notebase has ‘items’ and ‘grids’, an item is the basic unit of information, it has a title, a document pane which may or may not contain a document and it has a set of meta-data.
    • A grid is just a table of items, it is a filter which shows those items that meet the requirements for membership of that grid. It can be thought of as a database query.
    • Items exist independently of grids and is possible to have an item which doesn’t appear in any grid.
    • Each grid acts like a two pane organiser the navigation is simple as the items in the grid can be arranged into a hierarchy, so an item can have a number of ‘child’ items and this list may be expanded or collapsed just like a two pane organiser.
    • A grid can have a simple ‘grid source’ which is just a flag to say that the item is a member of that grid, all items with the flag set appear in the grid (this is the default). A grid may also have a ‘custom source’ which is an SQL SELECT statement or the name of an existing flag, all items meeting the conditions of this statement are included in the grid. This is similar to inline queries in ConnectedText or saved searches in MyInfo. Setting a Custom Source field for a grid can be a little complex for people who are not familiar with SQL (like me, but I am learning).
    • The contents of a grid like this get updated automatically when any item is changed.
    • Tagging in InfoQube has recently been updated to have a hierarchical tagging system and it has become extremely useful. Simple AND/OR type selections are very easy to do via the ‘Live Search’ pane. If more complex searches are required then a grid with a ‘custom source’ may be used and the criteria for selection can include Tags. InfoQube has very powerful search facilities.
    • This text does not represent all the sophisticated features provided by InfoQube, it barely scratches the surface, but suffice to say that all four of the facilities necessary for finding your information are well represented.
  • Big Data and the underlying file structure
  • InfoQube
    • InfoQube uses an SQL database as it’s storage mechanism and there is no need for the user to save the document as it is continually updated on the disk.
    • I was not able to carry out the load test as there is no way to bulk import text files, each one would have to have the text copied and pasted individually. So no information is available at this time.
    • InfoQube can open multiple notebases simultaneously however each one is opened in a separate instance of the program.
  • Transclusion & Linking
  • InfoQube
    • InfoQube is very flexible with respect to the layout of trees. Documents (items) can appear in multiple places in a tree and in multiple trees. Also the links to those documents are duplicated so if you add a child item to one instance it is automatically added to all instances of that document. This is transclusion done correctly.
    • InfoQube can generate Universal Links to content within InfoQube, you can link to various things within InfoQube like the Calendar, the Surface (a sort of mind map thing) or a specific document (but not to a position within that document).
    • I have placed an icon on one of the toolbars to generate a Universal Link to the current item. InfoQube didn’t have a suitable icon but InfoQube has an icon editor built in so you can roll your own.
    • InfoQube can link to files, folders, e-mail addresses, web pages and Universal Links generated by other programs.
  • Screen Presentation
  • There is a trick which can be applied and it works with some programs but not with others. Pierre Landry the developer of InfoQube told me to try setting the ‘High DPI scaling override’ to see what happens.
  • In Windows Explorer, right-click on infoqube.exe > Properties
  • The results were spectacular, but not just for InfoQube.
  • InfoQube
    • InfoQube is one of the most configurable programs I have used, except that it doesn’t support themes. Panes can be viewed and arranged on the screen in virtually any configuration. Panes can become floating and may be placed on a second monitor. You can also dock panes into various sections of the main window. This program is extremely flexible.
    • You can make your own toolbars or re-configure the existing toolbars, you can re-configure the menus. Although InfoQube has a very dense user interface this may be simplified somewhat by taking out the bits you don’t need.
    • InfoQube also had a problem with fuzzy edges.
    • Applying Pierre Landry’s high DPI tip sharpened up the edges. InfoQube now renders at the full native resolution of the high DPI screen and the text is incredibly sharp even at low point sizes. It now has stunningly sharp graphics.
    • The editing experience is excellent, my preference is to have the document pane taking up the full screen just above the laptop keyboard with the rest of the program full screen on the second screen.
    • The ability to set a .CSS file for each individual document is also really good.
    • In InfoQube you can use Internet Explorer 11 mode to render the HTML documents, this means that the .CSS files can include Linear Gradients and drop shadows. This may seem like a novelty but it is actually useful. Having a vertical gradient as the background colour to a page gives the user an unconscious visual cue as to how long the document is and where they are in the document.
    • The result is stunningly sharp and clear documents with excellent formatting in a WYSIWYG editor in a full screen almost distraction free view without the effort of having to format everything being edited. What more could one ask for?
  • Ease of Use
  • InfoQube
    • This program is so packed with features that the user interface is very dense. Until you become familiar with where to find things there are many times when you feel lost, this is the same for any complex program but perhaps InfoQube is a bit more complex than the average complex program.
    • Once you have learnt the basics of InfoQube the learning curve becomes less steep especially when you find out how to re-configure the user interface. But the problem is that the first part of the learning curve is especially steep for someone who is unfamiliar with the program and this is a big barrier to new users, but if you persist the rewards are well worth the effort.
    • Re-configuration to place the commands you need where you can find them is essential in my view. Different users will want different configurations but one of the beauties of InfoQube is that it is so configurable.
    • Also there are one or two concepts which you need to learn which will make everything else fall into place. Like the relationship of items to grids.
    • Unfortunately the documentation lags behind the actuality of the program because of the pace of development and the vast amount of material there is to cover.
    • It took a while to get my mind around InfoQube, I am still learning and there is still a long way to go.
  • Conclusions
  • As far as a comfortable writing environment goes InfoQube gets my vote.
  • Also for a reading environment InfoQube with it’s great .CSS files with the linear gradients and razor sharp graphics also gets my vote.
  • Navigation has to be a joint first for three of the four programs, InfoQube, MyInfo and Right Note. Navigation in ConnectedText is somewhat different to the other three and it takes a different mindset to become good at.
  • InfoQube has the best Favourites list, but only if you put the Favourites list into a grid. If you don’t know this trick then the best Favourites list is in ConnectedText.
  • The new tagging system in InfoQube is just as effective as the Categories system of ConnectedText although this may change as the tagging system in InfoQube is still being developed. So as far as tagging goes at the moment it has to be a joint first between ConnectedText and InfoQube.
  • Overall as far as retrieval goes, taking everything into account I would say it was a joint first between ConnectedText and InfoQube.
  • If you have vast amounts of data then the first prize must go to MyInfo but the tests on InfoQube could not be performed because of the inadequate import facilities.
  • Transclusion & Linking is difficult but it has to be a joint first between ConnectedText and InfoQube.
  • As far as the presentation on screen goes InfoQube with it’s razor sharp graphics and detachable floating panes is without question better than the other three.
  • Other Factors
  • The pace of development for InfoQube is frenetic. In the last six months it has acquired Universal Links, CSS sheets for the Document pane, Google Calendar synchronisation (both ways) and a hierarchical tagging system.
  • The bottom line
    • Taking everything into account if I had to choose just one program from the four and give up the others I think it would be a close run decision between ConnectedText and InfoQube.
    • They are very different programs and each one does things that the other cannot but these are mainly the features I don’t use. For example, all the project management and Gantt Chart stuff in InfoQube and the named blocks and all the CAQDAS stuff in ConnectedText. Looking at the features I do use the capabilities seem fairly similar. But they are very different programs.
    • InfoQube however is under rapid development by a developer who listens to the users of his program and tries to provide them with what they want. In one sense this is bad because it has led to a vast jumbled mish-mash of features which take some time to comprehend, and it leads to a complex user interface. In all other senses this is a good thing because everyone is getting what they want. It really is ‘everything and the kitchen sink’, whoever heard of an icon editor in a note taking program? But on the other hand I did find it useful to have a built in icon editor in InfoQube when none of the existing icons met my requirements.
    • Pierre has tried to make InfoQube everything to everyone and on the whole he has succeeded. It is a very open ended program which the user can adapt to solve many different problems.
    • Taking everything into account if I had to choose just one program from the four and give up the others I think it would have to be InfoQube.

DragonGem

2018/10/25 13:58

Permalink

Pierre, I must have caught you in "down moment"... sorry for diverting you from more pressing priorities.

Interesting approach with an extremely granular result! I keep forgetting that each [Item] can basically be a document (paragraph at least) by itself. I'm sure that I will find a good use for this technique somewhere along the line. So much of today's article publishing and blog postings have such a "stream of consciousness" quality that filtering down to just the core references with the ability to selectively add some additional context [Items] will come in handy.

My objective revolves around breaking thoughtful, well-structured articles like Paul's (with discrete Subjects and Topics) into more manageable chunks that are easy to manipulate. Think about it like combining Word's Outline Mode with its Collapsible Headers feature but throw in Filtering, Tagging and the ability to view, compare and edit multiple Topics simultaneously. I'll send you my example when I'm done with it.

I've already decided my parser should also include an option to break on a specified Word count (feature creep already).