After various experiments the best way I’ve found to get Scrivener content into a MarsEdit blog post is to complete to HTML then copy/paste the rendered HTML into MarsEdit.
Everything else messes up paragraphs.
After various experiments the best way I’ve found to get Scrivener content into a MarsEdit blog post is to complete to HTML then copy/paste the rendered HTML into MarsEdit.
Everything else messes up paragraphs.
A public Facebook group featured the best review of dropper posts I’ve read, by TB R-A. Reposting here so it’s not lost...
I’ve owned the RockShox Reverb, Fox Transfer, Specialized Command, Bontrager whatever it is called, and 9 Point 8 Fall Line. Of them all the Fall Line has had the best actuation and has been the best performing.
The Reverbs are a pain because they need bled. The bleed process is easy enough, but still a pain. Also, if the line would be damaged or cut out on a long ride you’re SOL. You can’t fix it in the field.
The Transfer worked well enough, but it makes a sucking noise at the top and bottom and drove me nuts.
The Specialized was a bit finicky, I don’t remember much more about it. The Bontrager was OK, but not nearly as nice as the Reverb in terms of quality or performance.
Right now I have the 9 Point 8 Fall Line on both of my bikes. Performance has been flawless and you can easily repair it on the trail should you need to. The upgraded lever made by WolfTooth is really nice, good ergonomics and you can just replace pieces of it should something break (which happened when I loaned my bike to someone). You can release the rail from the head on one side to access the air valve, it’s pretty slick. So, the Fall Line is my choice. I replace the seals annually on it, which takes about an hour. Aside from that nothing really needs done.
Dropper posts are expensive still, though the cheap clones are emerging. I like the idea of the wireless rockshox, but it’s $800 for now. The Bontrager is less costly than the 9 Point 8.
I’ve now bled my front and rear 2012 Avid XX brakes. I did the front first and got an excellent result. I did back twice, the first time it sucked the 2nd time it was good but not as good as the front. Tricky business! You really need to follow directions fairly precisely.
Want to get some quick notes out here mostly for my use. Maybe later I’ll fill in the rest.
I don’t think anyone is terribly interested in this, so I’m not going to provide a lot of detail, but if you want to know more just email jgordon@kateva.org.
So why am I writing anything? I’m writing because this thing has worked very well for years. Alas, nobody at work appreciates the simple genius of how it works. So here I can vent.
I ignored SharePoint’s clumsy native blog. I built my blog off the SharePoint announcement list which has basically worked the same way for 10-20 years. Like every list in SP it has an RSS feed (deprecated in newest versions of SP alas) and email notifications with user controlled update frequencies. So the subscription side works fine.
I added a couple of fields to the basic list. One is for tags. Tags let me create topical views of the list by creating SP views that filter on tags.
The other field (and this worked better than I imagined) was to create my own PUBLICATION_DATE field and sort the blog (list of announcements) by publication date. The default value is the date created — but I can edit the publication date without changing the post URL!
The last is wonderful. Instead of having to point to an old post and perhaps add some new additions, I just update the old post and revise the publication date to current. SP regenerates email notifications to subscribers, updates the RSS feed, and the view shows the updated post at the top — but the URL is unchanged so links don’t break.
It’s really simple, it’s worked very well for 2-3 years, it should work indefinitely. I wish other blogs worked that way — let me revise publication date, sort by that date, don’t break links.
PS. Why SharePoint? Because where I work that’s what we have. We aren’t getting anything better.
For years I’ve launched gmail and gcalendar on my Mac using apps in my user/application folder I’d named gCal.app and gMail.app. I loved them. Not only did they let me go directly to gCal and gMail from spotlight they also opened in my preferred user profile rather than the last user profile I’d used.
Today they crash on use.
Intermittently I got this error message:
Default pjkljhegncpnkpknbcohdijeoejaedia quit unexpectedly.
That looks like malware, but apparently pjkljhegncpnkpknbcohdijeoejaedia is an undocumented internal Google Chrome extension. (Very cute Google.)
After a bit of research I believe those apps were known as Chrome apps, and years ago Google supported creating them on the Mac. Google ended support in 2017. I think the apps lived in a Chrome folder that was removed, but on my machine they survived because I’d moved copies into a user folder.
Today my extended use luck ran out. Now they crash [1] under Chrome 75.0.3770.100. I suspect my Mac was updated today or yesterday, this version of Chrome was released June 18.
I REALLY miss the desktop app functionality. “Progressive Web Apps” were supposed to replace it but I don’t think they happened [2]. I might try this 2018 tip to see if I can restore it.
Or maybe the next release of Chrome will fix the crash :-).
- fn -
[1] sample start of log
Process: app_mode_loader [1087]
Path: /Users/USER/*/gCal (jfaughnan).app/Contents/MacOS/app_mode_loader
Identifier: com.google.Chrome.app.Default-ejjicmeblgpmajnghnpcppodonldlgfn-internal
Version: 4.5.6 (2564.97)
Code Type: X86-64 (Native)
Parent Process: ??? [1]
Responsible: app_mode_loader [1087]
User ID: 502Date/Time: 2019-06-29 14:33:12.397 -0500
OS Version: Mac OS X 10.14.5 (18F132)
Report Version: 12
Anonymous UUID: C285F89D-D3A8-7245-0199-81B760782A83
Time Awake Since Boot: 2100 secondsSystem Integrity Protection: enabled
Crashed Thread: 0 Dispatch queue: com.apple.main-thread
Exception Type: EXC_BREAKPOINT (SIGTRAP)
Exception Codes: 0x0000000000000002, 0x0000000000000000
Exception Note: EXC_CORPSE_NOTIFYTermination Signal: Trace/BPT trap: 5
Termination Reason: Namespace SIGNAL, Code 0x5
Terminating Process: exc handler [1087]Application Specific Information:
dyld2 modeThread 0 Crashed:: Dispatch queue: com.apple.main-thread
[2] No, didn’t happen on Mac. Over a year late now:
Progressive Web Apps on Desktop
Progressive Web Apps now work on the desktop, including Chrome OS and Windows, with support for Mac and Linux coming soon.
For Google macOS is not a priority platform.
As recently as 2018, Apple had stopped supporting multi-select in iCloud Photos.web. It was really annoying.
Today it works. I don’t know when this was fixed. It didn’t get a lot of attention in my streams.
It’s a big feature for me. Now if I want to add photos from my wife or daughter I can browse to their iCloud accounts, select from Photos.web, and download. Much easier than switching to a user account, running Photos.mac, waiting for it to sync, exporting to a shared account, etc, etc.
One odd thing, the file “Modified” date is “Tomorrow at 12:15am”. Presumably a server time zone problem….
There’s a well known old-device bug with Screen Time. Once you’ve setup up Screen Time for a child’s device it will always show up under their Apple ID — even after you wipe the device.
So my daughter’s old iPhone 6 showed up under her Screen Time Apple ID — even after it had been wiped.
Today I changed the Apple ID for that device and enrolled it in Screen Time under a new Apple ID. Then I went to look at her Screen Time device list. Lo and behold — the old device was still there, but now its device name changed to match the device name it had when I reenrolled it with the new Apple ID.
So now one device shows twice in my remote Family Screen Time, once under my daughter’s device list, once under the new Apple ID.
I’ve read that Apple’s Screen Time was a “rush job”. Looks like the dev is doing Screen Time by storing a device identifier — maybe a Serial Number and the device name used at enrollment time. They should have used a combination of Apple ID and device identifier but they used device identifier alone.