Forum Posts Feed
Text expander and clipboard enhancer tools
< Next Topic | Back to topic list | Previous Topic >
Pages: ‹ First < 9 10 11 12 13 14 15 16 >
Posted by Dr Andus
Dec 3, 2013 at 06:04 PM
Dr Andus wrote:
PhraseExpress wrote:
>>Video demo with a 22,000
>>words library:
>>http://youtu.be/PpH23B_qqso
>
>1) I didn’t have any problems with importing with PhraseExpander so far,
>so I’m not sure what’s happening in your example.
Actually, the import problem in your video might have to do with the fact that the CSV file you are trying to import has different column names than the ones you use in PhraseExpress. So rather than just clicking on the “CSV File” button to import, click on the little arrow next to it, which tells you what the column headings need to be. Then just change the names of the column headings in Excel, save, and try importing it again.
Posted by PhraseExpress
Dec 3, 2013 at 06:25 PM
You can see from the video that it actually recognizes the contents properly as shown in its preview window (Direct link to that moment: http://youtu.be/PpH23B_qqso?t=2m2s). So, everything is good with the file format. However, the actual import seems to take forever.
You may try the file we used (not our content, use at own discretion):
http://www.file-upload.net/download-8359470/22.000-English-words.csv.html
Have you ever imported a larger database? If you managed it, do you still find the “Smart"Complete feature useful?
Please don’t fool yourself by sticking to abbreviations you are used to. Rather look at the CSV file itself, pick a random word and then try to catch it with the “Smart"Complete feature. Observe yourself, how much time your brain needs to analyze the popup and which steps you undertake to narrow it down.
It would be interesting if you would make a video about it.
Posted by PhraseExpress
Dec 3, 2013 at 07:39 PM
Dr Andus wrote:
PhraseExpress wrote:
>1) I didn’t have any problems with importing with PhraseExpander so far,
>so I’m not sure what’s happening in your example.
It’s about the database size. Many users want precompiled sets of phrases/words to get started quickly.
>2) But, I’m using PhraseExpander’s SmartComplete feature very
>differently from your example. I create my own phrases incrementally and
>strategically. I only need very specific frequently used terms.
>Currently I have about 800 phrases.
Indeed, this is a completely different and pretty much time-consuming approach. Our customer feedback tells us that not many users are willing to entertain such enormous efforts. YMMV
>4) Your video played too quickly, so I wasn’t able to figure out your
>algorithm, but it seems to work quite differently from PhraseExpander’s,
Nope, while it is not the same, it is quite similar.
>as there seem to be a lot of not-so-relevant matches.
Haha, you got it. This is EXACTLY the point. :-)
The video shows that such algorithm can’t cope with large word libraries. It’s simple logic. If you have 100 words beginning with “co” and ending on “ly” it becomes difficult to narrow one particular item. That is the whole point of our concern.
Please try to import the file and tell me if you are still happy with the feature.
>Thanks, sounds like an interesting feature. But the big difference that
>comes across between your approach and PhraseExpander’s approach is that
>you seem to want to predict what I want to write,
This is correct for our text prediction feature but this is not what is shown in the video.
The video shows a prototype of our “Smart"Complete approach which is also based on a static dictionary. You can also change order, positions, priority, etc. You won’t be able to analyze how it works as the algorithm has some floating parameters. It was just to demonstrate a popup that is constantly flashing with every character.
We will polish our approach and it would be exciting to have you test it - You seem to be exceptionally fascinated with this particular feature and it will be fun for you, too because it has some special sauce that will make you say “HOLY COW!”. ;-)
Posted by PhraseExpress
Dec 3, 2013 at 08:11 PM
>I type the abbreviation, PE expands it. THIS time I do not want that, so
>I undo the expansion.
Got it. This is the point where our idea and your use scenario starts diverging:
PhraseExpress: “OK, he wanted to type the abbreviation but didn’t wanted the expansion because he undo it. So, when he types the abbreviation again to return to the state right before the text expansion, I will ignore it this time.”
Franz: “I entered the abbreviation but NEITHER want the abbreviation nor the text replacement anymore on this occasion”.
If this is correct, then indeed, the unwanted Autotext detection is not for you and you can/should disable it.
>Even more annoying is: That the third time PE changes behaviour again.
If you undo a text replacement for three times during one session, PhraseExpress assumes that you actually do not want this particular text replacement at all and offers(!) you to remove the item from the library. This behavior is connected to the “unwanted Autotext detection” and can be disabled.
PhraseExpress best intention is: Why would anyone undo a text replacement for three times in a row while still wanting to have it in the library?
>Don’t you understand that a user wants software to work this way: When
>the user does A, he wants the software to reliably do B. He doesn’t want
>it to do C and then the next time again B. How do you expect your users
>to rely on the software, if it does strange things (“by design”)?
Please put yourself into Joe Average’s seat to understand the motivation of the feature:
Joe types “FBI” and PhraseExpress expands it to “Federal Bureau of Investigation”. But on this specific event, Joe wants only the short form, so he undo the expansion and types “FBI” again.
Any other Autotext software we are aware of would enforce the expansion again. Joe would probably say “What the heck is going on? Haven’t I just removed this ******-****** stuff. I don’t want the expansion! Leave me alone. I want to write “F B I” and nothing else. Dammit! Get out of the way!”.
Joe would probably even use stronger words after the third time of undoing the unwanted text expansion and throw his computer out of the window.
However, we don’t like being hit by computers while walking on the street. So, with PhraseExpress, Joe simply undo the expansion and re-type FBI and Joe is happy because PhraseExpress holds back this time.
The next time, Joe might want the expansion again and PhraseExpress behaves as expected. If he doesn’t like the expansion at all, he undo again and again and PhraseExpress reacts on it and suggest to remove this expansion from the database,
—
I have difficulties to understand why you enter an abbreviation, undo the expansion and then surprisingly completely give up entering the abbreviation again (as this is what you did firsthand). Why have you entered the abbreviation in first place at all?
Perhaps we really miss out a use scenario here? This would be really interesting to learn.
Posted by Franz Grieser
Dec 3, 2013 at 08:44 PM
>Perhaps we really miss out a use scenario here? This would be really
>interesting to learn.
Oh yes, you do.
>Joe types “FBI” and PhraseExpress expands it to “Federal Bureau of
>Investigation”. But on this specific event, Joe wants only the short
>form, so he undo the expansion and types “FBI” again.
That’s the point. When after the expansion Joe presses CTRL+Z to undo the expansion, he does not need to retype “FBI” because it’s already there. CTRL+Z undoes the last action, which is the expansion.
That’s how it works in Word, in Outlook and in OpenOffice/LibreOffice - and I guess in almost any Windows application.