Conversion of large websites into WA
I would tell anyone with big websites that the conversion is the most difficult part of going to Wild Apricot, especially when there is need to convert many pages. The conversion process was quite painful. We had php pages with very simple and clean html code. Our subcontractor apparently copied and pasted from the displayed webpage instead. This resulted in poor conversion of links and images. Where we had had had a central repository folder for all the images on the old website, we ended up after the conversion having images attached separately to each page and in many cases links going back to the soon-to-be-defunct website. After the contractor was done, our volunteers had to recreate the central repository of images and revise every page manually and redirect links to that common images folder or to pages on the new website and away from the old website. There must be things WA could do or recommend to make this conversion process easier.
Alex Sirota commented
A few comments here -- export functionality is a must, as copy and paste is rather cumbersome. An import of the WordPress XML files could be a nice addition.
The main issue is that the new URL format must be established during the import to ensure link names are consistent in the new ** site.
Having an import from WordPress feature would go a long way to fixing this.
Karen Halbert commented
I absolutely agree with your statement: "That said, Dmitry, it would be great to have some sort of link checker built in - nothing super fancy, but something that identifies broken links across the site (if you delete or move something for example). Equally useful if you're looking to remove images from the repository - would be great to know where an image is being used before you delete it!"It would be extremely useful to have both items.
Dmitry Buterin commented
Kerrie-Anne - thanks, noted.
Conversion is difficult, but no more difficult than conversion between any other two systems - you just have to plan your migration process carefully and ensure you have someone who is up to the task.
I found that in the case of WA, there is a lot of manual work involved (I've yet to come across anything that automates the migration process particularly well), but if you were using a central repository on your old site, you can upload that structure to the WA tree via webdav this should have been one of the first things done - this essentially recreates your central repository folder, making changes to image references consistent.
From there, even if you had cut & paste from the rendered/display site, it would be a simple matter to go into the html code for relevant pages and do a "find & replace" for any updates to the file path
Of course any contractor worth their weight will have tested all the work they had done before handing over to you - they could also have run some simple link testing reports across the site as well to ensure they had migrated everything correctly.
That said, Dmitry, it would be great to have some sort of link checker built in - nothing super fancy, but something that identifies broken links across the site (if you delete or move something for example). Equally useful if you're looking to remove images from the repository - would be great to know where an image is being used before you delete it!
If a built in link tool is not an option, then perhaps an interface so that it can be hooked in to some of the tools built for this purpose that are out there.
Dmitry Buterin commented
Tom, thanks for bringing up this issue.
I would appreciate comments from other users about their experience converting their website to Wild Apricot - what were the bigger challenges and how do you think we could handle these better in our software.