Data Migration

[rdprdp01]

I have been working closely with Mostafa about data migration options. Here is where we landed. Before I spend the money to make it happen, I need your guidance and direction. It is only worth doing if you see that it will yield a return for the Members:

We can “screen scrap” the first post of every thread on the prior Forum. That scrape will include all texts and graphics from the initial post but none of the responses. We can organize each Thread by year under a single user (Archive). We can insert “Archive” at the beginning of the scraped threat to denote that it came from cb1100forum.com. Because the old site is unstable, it will take several days to complete the port but it will finish. Please provide your thoughts and opinions about this option in this thread so everyone can have the benefit of your insights.

If you find this option lacking, I can manually port over all Posts for specific threads manually, but I can do only so much during the course of the week. You could join me, too, but I assure you the content is interesting but the process is tedious.

What is best for the future of this Forum? Please advise.

[Cormanus]

My 2 cents ...

It seems to me problematic just to bring the first post of most threads as much of the substance actually occurs in later discussion. For example, the first post of the very long discussion about the low/idle issue would not be much help. It was the discussion that produced a variety of solutions and suggestions.

Perhaps the first post of 'how to' threads would be good to capture as it's to be hoped that's where the critical information is.

Because I think the inane RLETs thread says a good deal about the character of the old forum, I've made PDF files of all the pages. I'll either upload them here or put them somewhere on the web and link to them. However, it will be much less satisfactory to read than a thread on a forum. I have thought to do the same with Ferret's history threads. That way the information is captured somewhere.

I'd be interested in other views.

[rdprdp01]

I do not mind porting threads manually. It is how I earn my place at the table and show my respect. I just need some guidance about what to port and what not. I think you and I are in violent agreement about where to start.

As long as the old Forum remains up and running, I believe I can port the most important threads before December, but given my other obligations, it will take until the end of the year to complete.

Let’s see where The Ferret and others land. There is also nothing stoping other Members from recreating their greatest hits in the new Forum, too.

David

[kennyw]

My friend Gary (also user name Gary on the old forum) recommends looking at Beautiful Soup and Scrapy for the data migration automation, he thinks you could automate porting the data with those and a navigation script. He automates website stuff for a living (I know nothing about this stuff myself).

https://beautiful-soup-4.readthedocs.io/en/latest/

https://docs.scrapy.org/en/latest/

[rdprdp01]

Thank you, Kenny! Will check it out. I was surprised that there is not more enthusiasm for porting data. I will give it a few more days and see if there is more demonstrative support. What do you think? Should we just start fresh or port data? David

[Cormanus]

I'm all for porting data from the old forum if it is relatively easy to do and can capture entire threads along with some of the meta data such as who wrote the post in the thread captured. Just grabbing the first post in a thread will yield some good stuff, but will loose anything helpful that came afterwards. For example:

1. EmptySea's ride reports tend to be written episodically so capturing only post 1 would mean losing the gold that is the rest of the tale. 
   
2. Capturing a first post that asks what oil to use in the CB1100 but doesn't provide the answers is, possibly, not all that helpful.
   

If, on the other hand, one can automate grabbing entire threads, let's do it.

[rdprdp01]

I agree with you. Unfortunately, the only way I know to do it with 100% reliability is to manually select and copy one thread at a time. Let me post an example I did after this post. I desperately want to (1) keep great content, (2) shed not so valuable content, and (3) automate the process with a tool or custom code. Here is what a manual block and copy looks like.

[max]

Thanks for all your effort David, a beast of a job, good to see and thanks for letting me join ( what a mouthful )

Not too bad with bikes mechanically but embarrassingly bad at software, too old and very capable messing such things up!

Cheers max.

[rdprdp01]

Sample Block & Copy

The Challenge: A user (the first one will be yours truly) will pick a location, landmark, geographical feature, etc. to serve as the backdrop for your photo. Users will then post photos meeting the requirement in this thread for our ogling pleasure. The first user to submit a qualifying photo chooses the backdrop for the following month's challenge.

The goal: To encourage users to get out and ride and to document those rides with photos and, most of all, to have a little fun with the whole process.

The Rules:
1. The CB1100 must be the featured bike in the photo. Caption the photo with the date and the location.
2. Feel free to alter your photo to make it look better, but don't cut and paste into another photo just to qualify.
3. The photos should be new photos, not something you had on file and especially not something you posted elsewhere in this forum.
4. You may post multiple qualifying pictures in a month, but the background has to be different (ex. you can post two pictures of your bike in front of a train station, but they have to show two different stations)
5. Not really a rule, but don't stop trying to get a picture for a particular month just because someone beat you to the first post for that month. We want to see lots of pictures. 
6. The same user cannot pick the photo subject on consecutive months. If the first photo comes from the same user that posted the first photo in the prior month, the second user posting a photo will get to pick the next background subject.

Truly, 6 rules are probably 5 too many. Just go with the spirit of this challenge and get out and ride your motorcycle!

---

January Challenge (from EmptySea):

Let's see your CB1100 in front of a statue. Ready? Go!

---

MTC

"If you can’t go home again, at least you can enjoy the ride” — Peter Egan


2013 CB1100 non-abs
2013 CB1100 abs

(This post was last modified: 01-16-2017 09:06 PM by EmptySea.)

01All righty, I'll play.
Rode out in the rain (and 45 degrees) for this collection of statues. There are three bronze gods chained to a wall in an out of the way office park in Falls Church, VA. I don't know why they are there or who they belong to. They are very intricately detailed and about 10 feet tall. You can't see them from the road and I had to ride up several sidewalks to get the shot. But that's just how much I care about sharing with you guys :-)
[/url]

Bonus shot. This little guy is right there, too.
[url=http://s1331.photobucket.com/user/inhousebob/media/IMG_1065_zpseri5aucy.jpg.html]


17-Those of us under a blanket of snow can't participate, but I'm sure many will still enjoy the contest vicariously.

---

Still rockin', rollin', and ridin' after all these years
'67 BSA 441 VR, '70 Triumph Tiger 100, '02 Honda CB900F, '06 Triumph T100 Bonneville, '14 Honda CB1100A


That's awesome Bob! Those statues are one thing, but the donkey statue too...quite an eclectic collectio

Great pictures, Bob. And thanks for the extra effort in obtaining them.

---

MTC

"If you can’t go home again, at least you can enjoy the ride” — Peter Egan

Moderator


Good work, Bob. 
And a good challenge MTC.

---

It is not perfect, but it does convey much of the impact of the oringinal post. I will keep working to find a more complete option. So far, I am striking out, sadly. Good night. David

[Cormanus]

Thanks rdprdp01. I can see the dilemma. In my view, block copy and paste is too tedious a task to take on and requires too much manual intervention to capture the metadata such as the name of the poster. Not much other metadata is required.

Your example above highlights the problem, I think. You can figure there are a couple of people posting, but who and where the breaks are is not immediately obvious.

Like you, I'd welcome other thoughts on this topic.