3 ms·

Show HN: Factonaut. Collect historical events and create chronological overviews

7 points by Archelaos 6y ago

Archelaos 6y ago
Author here. When COVID-19 hit the tourist industry, for which I was coding, and my assignments were deferred, I suddenly had a lot of time to work on a side project of mine: 'Factonaut', a Windows desktop application that allows to compile historical events in a knowledgebase and create chronological overviews. Although I still have a huge mass of unimplemented features on my to-do list, I finally took the courage to release a minimum viable product. Compared to similar software, it does three things quite well: it copes with Gregorian as well as Julian calendar dates, it allows to use date (and time) values of different magnitudes (from the second up to billions of years) in one chronology, and it does a decent job when using LaTeX as an output format. Apart from that, it lacks many of the features and ease of use that other applications offer. However, it is difficult for me to decide what to focus on next. I would especially appreciate feedback from people who find the Julian/Gregorian feature useful, about what else matters for them most. You may reach me via the email adress in my HN profile or on the Factonaut Web-site.
ksaj 6y ago
This is great. As a hobby, I have a text file that I fill with interesting (to me) historical records, sorted by date. The reason I do this is because I enjoy seeing how certain things already known about and perhaps taken for granted, relate to other historical interests that happened simultaneously. As well, it can show interesting timelines, like the discovery of something followed by its eventual implementation, and the side effects thereof.
mncharity 6y ago
Perhaps create some "knowledgebases" by bulk scraping and linking? For an old project ("if days are dots, then several centuries fit on a screen, and you can mouseover history"), I scraped Wikipedia's On This Day[1], and linked into US Library of Congress scans of newspapers like the New York Tribune (1866-1922)[2], and Google's scans like LIFE magazine[3]. For illustration, here is Dec 13-ish for OTD, LIFE 1950, and Trib 1920 and 1890.[4] [1] https://en.wikipedia.org/wiki/Wikipedia:Selected_anniversaries https://en.wikipedia.org/wiki/Wikipedia:Selected_anniversari... [2] https://chroniclingamerica.loc.gov/lccn/sn83030214/ https://chroniclingamerica.loc.gov/lccn/sn83030214/ from https://chroniclingamerica.loc.gov/ https://chroniclingamerica.loc.gov/ [3] https://books.google.com/books?id=gFUEAAAAMBAJ https://books.google.com/books?id=gFUEAAAAMBAJ [4] Dec 13-ish OTD https://en.wikipedia.org/wiki/December_13 https://en.wikipedia.org/wiki/December_13 LIFE 1950 https://books.google.com/books?id=vEsEAAAAMBAJ&printsec=frontcover&source=gbs_ge_summary_r&cad=0 https://books.google.com/books?id=vEsEAAAAMBAJ&printsec=fron... Trib 1920 https://chroniclingamerica.loc.gov/lccn/sn83030214/1920-12-13/ed-1/seq-1/ https://chroniclingamerica.loc.gov/lccn/sn83030214/1920-12-1... Trib 1890 https://chroniclingamerica.loc.gov/lccn/sn83030214/1890-12-13/ed-1/seq-1/ https://chroniclingamerica.loc.gov/lccn/sn83030214/1890-12-1... Jefferson Republican 1840 https://chroniclingamerica.loc.gov/lccn/sn86053954/1840-12-11/ed-1/seq-1/ https://chroniclingamerica.loc.gov/lccn/sn86053954/1840-12-1...
- Archelaos 6y ago
  Author here. -- Your suggestion inspired two new items on my to-do list: - Support 3rd-party plugins for bulk additions of events (eg. from Web scraping). - Support customized dynamic links based on event dates to Web resources such as newspaper archives, Wikipedia articles, etc. Both are perhaps not top priority, but would be really nice to have.
mncharity 6y ago
It seems .fcntkb knowledgebase files are raw SQLite database files. For the long-term preservation of user work, perhaps something non-binary would be better, like JSON?
- Archelaos 6y ago
  Author here. Transforming the SQLite format to something else is IMHO not that critical in terms of long term preservation (= several decades). What is really important is that the semantics of the data format is documented and ideally standardized. If the project gets enough traction and the format stabalizes, I plan to go for it. I have also an item 'Export to CIDOC CRM' on my to-do list. CIDOC CRM is a standardization effort for cultural heritage data, backed by a lot of world's most prominent museums.[1] I am not sure that such an export could preserve every aspect of the Factonaut data model, but most of it. [1] http://www.cidoc-crm.org/ http://www.cidoc-crm.org/

7 points by Archelaos 6y ago

Author here. When COVID-19 hit the tourist industry, for which I was coding, and my assignments were deferred, I suddenly had a lot of time to work on a side project of mine: 'Factonaut', a Windows desktop application that allows to compile historical events in a knowledgebase and create chronological overviews. Although I still have a huge mass of unimplemented features on my to-do list, I finally took the courage to release a minimum viable product. Compared to similar software, it does three things quite well: it copes with Gregorian as well as Julian calendar dates, it allows to use date (and time) values of different magnitudes (from the second up to billions of years) in one chronology, and it does a decent job when using LaTeX as an output format. Apart from that, it lacks many of the features and ease of use that other applications offer. However, it is difficult for me to decide what to focus on next. I would especially appreciate feedback from people who find the Julian/Gregorian feature useful, about what else matters for them most. You may reach me via the email adress in my HN profile or on the Factonaut Web-site.

ksaj 6y ago

This is great. As a hobby, I have a text file that I fill with interesting (to me) historical records, sorted by date. The reason I do this is because I enjoy seeing how certain things already known about and perhaps taken for granted, relate to other historical interests that happened simultaneously. As well, it can show interesting timelines, like the discovery of something followed by its eventual implementation, and the side effects thereof.

mncharity 6y ago

Perhaps create some "knowledgebases" by bulk scraping and linking? For an old project ("if days are dots, then several centuries fit on a screen, and you can mouseover history"), I scraped Wikipedia's On This Day[1], and linked into US Library of Congress scans of newspapers like the New York Tribune (1866-1922)[2], and Google's scans like LIFE magazine[3]. For illustration, here is Dec 13-ish for OTD, LIFE 1950, and Trib 1920 and 1890.[4] [1] https://en.wikipedia.org/wiki/Wikipedia:Selected_anniversaries https://en.wikipedia.org/wiki/Wikipedia:Selected_anniversari... [2] https://chroniclingamerica.loc.gov/lccn/sn83030214/ https://chroniclingamerica.loc.gov/lccn/sn83030214/ from https://chroniclingamerica.loc.gov/ https://chroniclingamerica.loc.gov/ [3] https://books.google.com/books?id=gFUEAAAAMBAJ https://books.google.com/books?id=gFUEAAAAMBAJ [4] Dec 13-ish OTD https://en.wikipedia.org/wiki/December_13 https://en.wikipedia.org/wiki/December_13 LIFE 1950 https://books.google.com/books?id=vEsEAAAAMBAJ&printsec=frontcover&source=gbs_ge_summary_r&cad=0 https://books.google.com/books?id=vEsEAAAAMBAJ&printsec=fron... Trib 1920 https://chroniclingamerica.loc.gov/lccn/sn83030214/1920-12-13/ed-1/seq-1/ https://chroniclingamerica.loc.gov/lccn/sn83030214/1920-12-1... Trib 1890 https://chroniclingamerica.loc.gov/lccn/sn83030214/1890-12-13/ed-1/seq-1/ https://chroniclingamerica.loc.gov/lccn/sn83030214/1890-12-1... Jefferson Republican 1840 https://chroniclingamerica.loc.gov/lccn/sn86053954/1840-12-11/ed-1/seq-1/ https://chroniclingamerica.loc.gov/lccn/sn86053954/1840-12-1...

Archelaos 6y ago
Author here. -- Your suggestion inspired two new items on my to-do list: - Support 3rd-party plugins for bulk additions of events (eg. from Web scraping). - Support customized dynamic links based on event dates to Web resources such as newspaper archives, Wikipedia articles, etc. Both are perhaps not top priority, but would be really nice to have.

mncharity 6y ago

It seems .fcntkb knowledgebase files are raw SQLite database files. For the long-term preservation of user work, perhaps something non-binary would be better, like JSON?

Archelaos 6y ago
Author here. Transforming the SQLite format to something else is IMHO not that critical in terms of long term preservation (= several decades). What is really important is that the semantics of the data format is documented and ideally standardized. If the project gets enough traction and the format stabalizes, I plan to go for it. I have also an item 'Export to CIDOC CRM' on my to-do list. CIDOC CRM is a standardization effort for cultural heritage data, backed by a lot of world's most prominent museums.[1] I am not sure that such an export could preserve every aspect of the Factonaut data model, but most of it. [1] http://www.cidoc-crm.org/ http://www.cidoc-crm.org/