One caveat with the current approach is line breaks in bullet points. The 
current quick hack replaces line breaks with <br>, which is ugly to say it 
midkly. Any thoughts on alternatives here are welcome.

The issue with paragraphs in TW is also that the automatic paragraph 
handling is odd and a little inconsistent, and paragraphs can't contain 
lists so there will be a not so insignificant gap between them.

On Tuesday, May 5, 2020 at 1:19:56 PM UTC+2, Anne-Laure Le Cunff wrote:
>
> Oh, this is exciting! Thanks so much for looking into it, Saq.
>
> I think this is a case of needing to figure out what is most intuitive in 
>> terms of the end result the user would want, instead of what feels 
>> semantically the most correct. I suspect Roam users wont really want 5 
>> pages to turn into 100 tiddlers in TW, even if we did create a parent 
>> tiddler for each page that transcluded all the tiddlers created from the 
>> bullet points. 
>
>
> I agree with this. I think one page in Roam = one Tiddler in TiddlyWiki.
>
> The tricky thing is, do you turn the children into bullet points, 
>> paragraphs? You could even make that a global configuration option but then 
>> I suspect users would want to decide on a per tiddler basis... 
>
>
> For this one I disagree with Yoni. It should be bullet points in 
> TiddlyWiki as well, to stay as close as the Roam experience. It will feel 
> more intuitive. Very easy in TiddlyWiki to remove bullet points by 
> selecting all and removing the list styling.
>
> The tag handling is also difficult because Roam doesn't really have tags, 
>> I think what we are referring to as tags are just links. So should we 
>> really be trying to create tags from [[link]] or #link? Or do we assume 
>> that the links/backlinks that will be generated are the proper 
>> representation for this?
>
>
> Here I think, again in the spirit of staying as close as possible to Roam, 
> to not even consider the question of tags. Whether #Title or [[Title]] 
> these should all be converted to [[Title]] in TiddlyWiki. It may seem 
> extreme but I think that's the cleanest way to respect the original 
> structure from Roam and to keep things simple.
>
> On Tuesday, May 5, 2020 at 12:00:22 PM UTC+1, Saq Imtiaz wrote:
>>
>> @Riz: the number of nested children seems unlimited. In Roam the smallest 
>> semantic unit seems to be a bullet point, which is member of the childrens 
>> array in the JSON.
>>
>> I think this is a case of needing to figure out what is most intuitive in 
>> terms of the end result the user would want, instead of what feels 
>> semantically the most correct. I suspect Roam users wont really want 5 
>> pages to turn into 100 tiddlers in TW, even if we did create a parent 
>> tiddler for each page that transcluded all the tiddlers created from the 
>> bullet points. 
>>
>> The tricky thing is, do you turn the children into bullet points, 
>> paragraphs? You could even make that a global configuration option but then 
>> I suspect users would want to decide on a per tiddler basis... 
>>
>> The tag handling is also difficult because Roam doesn't really have tags, 
>> I think what we are referring to as tags are just links. So should we 
>> really be trying to create tags from [[link]] or #link? Or do we assume 
>> that the links/backlinks that will be generated are the proper 
>> representation for this?
>>
>> @yoni I understand your use case and requirements. My approach is to 
>> start off step wise and try to put something together that would be useful 
>> to others wanting to import from Roam as well. That may mean that there 
>> wont be one "plugin" that does what you need, but rather a sequence of 
>> steps/plugins working together. 
>>
>>
>>
>>
>>
>> On Tuesday, May 5, 2020 at 12:35:43 PM UTC+2, Riz wrote:
>>>
>>> I was having the same idea as Mark - treat each child as a separate 
>>> tiddler. Because that is how usually things are in TW5 - smallest semantic 
>>> unit and all. But I guess each child is a bullet point? 
>>>
>>> JSON mapping would be much easier if you have a definite structure. So 
>>> if the children doesn't have another level of children, it makes writing a 
>>> generalized importer much easier.
>>>
>>> On Tue, 5 May 2020, 13:49 Yoni Balkind, <ybal...@gmail.com> wrote:
>>>
>>>> Okay so this is looking promising! Thanks @Saq
>>>>
>>>> The import treats each child as a bullet point. This mimicks the way a 
>>>> page looks in Roam, but the idea of publishing on TW is that it reads more 
>>>> like a normal post, so I propose that you treat each child as just a 
>>>> paragraph (however nested children can still be treated as bulleted 
>>>> lists). 
>>>>
>>>> I now see an additional problem. This might be too unique to my own 
>>>> use-case to warrant you catering for it, but let me tell you anyway in 
>>>> case 
>>>> you think its a broad enough problem..
>>>>
>>>> Roam has no "tag" field per se.. Every internal link is a defacto tag.. 
>>>> So for example, the screenshot below shows a Roam note that has 4 
>>>> different 
>>>> internal links. For my purposes, the two links in the top line operate as 
>>>> tags. I use these tags to help me locate pages that I've tagged as 
>>>> "evergreen notes" and "ready for publish". 
>>>>
>>>> The 2 links near the bottom of the page are simple internal links to 
>>>> cross reference other posts. Roam doesn't know the difference between all 
>>>> of these links, I just use them in such a way that the ones on top are 
>>>> "tags" for my purposes. 
>>>>
>>>> Ideally, I'd want the import to ignore the two tags at the top. They 
>>>> are internal references that aren't needed in TW. I suppose IF you were to 
>>>> tackle this in your plugin, you could have a setting to ignore hashtag 
>>>> items. Or perhaps you could stipulate that should a user want to have 
>>>> certain lines ignored they should precede that line of text with "meta:" 
>>>> or 
>>>> something to that effect.. 
>>>>
>>>>
>>>> [image: Annotation 2020-05-05 095517.png]
>>>>
>>>> Final point, and again this might be my personal problem and not 
>>>> applicable to your plugin, but my JSON file has 1000 plus pages and I need 
>>>> a way to only import those pages that have certain tags (#ready for 
>>>> publish).. There is a probably some tool somewhere that would allow me to 
>>>> do this directly to the JSON file, but perhaps its something you want to 
>>>> consider for your plugin. 
>>>>
>>>>
>>>> -- 
>>>> You received this message because you are subscribed to a topic in the 
>>>> Google Groups "TiddlyWiki" group.
>>>> To unsubscribe from this topic, visit 
>>>> https://groups.google.com/d/topic/tiddlywiki/cMsvhrOgoBg/unsubscribe.
>>>> To unsubscribe from this group and all its topics, send an email to 
>>>> tiddl...@googlegroups.com.
>>>> To view this discussion on the web visit 
>>>> https://groups.google.com/d/msgid/tiddlywiki/925db46b-24ad-4846-ade3-b56637aa173a%40googlegroups.com
>>>>  
>>>> <https://groups.google.com/d/msgid/tiddlywiki/925db46b-24ad-4846-ade3-b56637aa173a%40googlegroups.com?utm_medium=email&utm_source=footer>
>>>> .
>>>>
>>>

-- 
You received this message because you are subscribed to the Google Groups 
"TiddlyWiki" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to tiddlywiki+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/tiddlywiki/e95d1c5f-d5f2-4582-a2b4-67b560e87939%40googlegroups.com.

Reply via email to