I’ve always used Instagram to promote my DJ hobby stuff. I know it is owned by (and integrated with) Facebook, which I actively dislike, but I always viewed IG as “Evil Adjacent” rather than “Pure Evil”. But at some point along the way, FB has now “streamlined” some things between the two platforms. The result is that if you want to use a scheduling app (like Buffer or Later) to queue your IG posts ahead of time, you need an IG Business Account that must be connected to a Facebook Page. You can’t just have an IG Creator account anymore, you must also have an FB account.
\nSo where does this leave folks who have no desire to be on Facebook?
\nAnswer: “no post scheduling for you”. And I know, if you are a casual user this is no big deal. If you are a content creator who has limited time to try and maintain a social media presence (which I am super bad at even with a scheduler), then this is a massive blocker. FB wants you all the way in.
\nMany folks have thrown in the towel and grudgingly set up their Facebook account. I get it. The fear that creators feel is that they will not be discovered and appreciated if they aren’t where the eyeballs are. And yes, many of their livelihoods depend on it. But I am increasingly becoming convinced that the big social media platforms always take more than they give.
\nMy DJ teacher wisely observed the the best way to gain a following is to start a mailing list. Plenty of services out there will help you manage your mailing lists, but the beauty of email is that you can always take that list with you. No matter what happens to the platforms you use, the mailing list is yours. I think about that a lot these days, and in the meantime, I’ve been slowly reducing my footprint on Meta.
\n“But I’m sure going to miss all of the ads…” – Nobody, Ever.
\n", "date_published": "2022-11-01T22:37:00+02:00", "url": "https://harrisonripps.com/2022/11/01/meta-problems.html", "tags": ["DJing and Music","Tech and Work"] }, { "id": "http://djharrisonripps.micro.blog/2014/08/26/os-x-to-fedora-the.html", "title": "OS X to Fedora 20: The User Experience, Pt. 2", "content_html": "Just tuning in? Have a look at Part 1 to find out how I installed Fedora 20 on a MacBook Air and why I am so interested in trying to duplicate or even surpass the user experience that I previously enjoyed with OS X.
\nCloud-based mail, calendaring and file synching have really levelled the playing field between the desktop OSes. It doesn’t matter if you are using OS X or Linux if you read your e-mail in a web browser. Cloud-based services are also the foundation of another important aspect of my day-to-day life: my “gadgets ecosystem”. Even though I’m spending most of my time on Fedora 20 now, my phone and tablet still run iOS. As you will see, this strongly influenced the way that I’ve leveraged the cloud in my system setup.
\nHere are the services that I needed to put in the cloud:
\nSome of these were no-brainers, others drove me to try and end up loving some interesting new services…
\nPeople don’t normally think of mail as a cloud service, but for most folks, it is. I could run my own mail server but I like the fact that my e-mail stuff is being handled by someone who’s full time job is e-mail, and that the service is configured for high availability. That said, part of the reason that I use a native mail client rather than just using my mail service’s web client is so that I can also have a local copy of all of the e-mail. This makes for faster searches and I always have those e-mails on hand if the mail service goes offline.
\nThat all said, mail is pretty much the original cloud service and IMAP keeps everything in sync for me, so problem solved.
\nIt may seem odd to think of an address book as being separate from a mail service, but in the cloud, they are. If you use Gmail, Hotmail, Yahoo mail, or any of the other popular free e-mail services, you never have to worry about this distinction. This is because Fedora and OS X both understand the concept of bundled internet accounts. You tell your computer “I have a Google account” and then your computer asks you which services to use from the bundled list.
\nBut I don’t use Gmail for mail, and furthermore I actually dislike Google’s approch to putting every last e-mail address and Google Plus profile in my address book without my consent. For a long time I used iCloud’s address book, but that’s a no-go from a Linux system. So, enter ownCloud.
\nownCloud could actually be the answer to a few of the questions in this post, but in my setup I use it expressly for my address book. This lives on an ownCloud server hosted on OpenShift. Advantages:
\nI have to grudgingly give a nod to Google on this one. By solving calendaring and then building a social network with built-in event scheduling, they’ve created a siituation where it is hard to use anything else. ownCloud supports calendaring, but scheduling events across different calendaring systems seems very spotty.
\nI threw up my hands on this one; too many folks that I need to share my non-work plans with are all on Google, so I have to live with Google for this as well.
\nApple’s “Reminders” app adds a lot of value beyond making lists. Probably the strongest feature there is the ability to geotag a task (e.g. Remind about picking up milk when I get to the grocery store). What “Reminders” doesn’t do is enable me to quickly create tasks from the Terminal, which is where I am working most of the time.
\nI’ve ended up splitting the difference on this one. If I need to geotag a task, I’ll create it in Reminders app on my iPhone. Otherwise, I have found an awesome task list solution in Todo.txt.
\nTodo.txt is great for a few reasons:
\nThere’s a blog post idea in my future where I talk more about why plain text file formats are so great, but for now, the takeaway is that todo.txt uses a winning combination of simplicity and optional cloud availability.
\nFinding an alternative to Notes.app can be dirt simple depending on what you want out of the app. At a minimum, you could simply designate a folder in your DropBox for notes and voila, problem solved. This is particularly true in Fedora 20 and OS X, both of which come with automatic file indexing.
\nHowever, I wanted faster navigation between notes and more targeted searching, and that led me to try out nvpy. This oddly-named Python of Notational Velocity makes use of global hotkeys and a very complete set of app-wise key combinations to provide a note-keeping and note-searching powerhouse. It can sync with SimpleNote’s API but it also does perfectly well writing Notes to text files, which brings us back into DropBox territory.
\nnvpy also runs on OS X, but nvALT looks a little slicker on OS X and can handle the same DropBox-synced note files as nvpy. There is one difference: nvALT will allow you to create notes with whitespace in the titles while nvpy will not, so get into the habit of putting underscores in the first line of the note.
The main benefit to this approach is that I can still get at those same notes from iOS. Elements for DropBox is a Markdown-friendly notes app that has no trouble displaying the notes that I create in nvpy.
\nAs an aside, ownCloud has a plugin for Notes, but the only way to see / use the notes is through the web UI or the native ownCloud apps. If I really wanted to solve this I’d go with the “just use text files” solution, but we get access from Fedora, OS X, and iOS with DropBox.
\nBy now I think you can guess which direction I took. With DropBox and ownCloud to choose from, I’ve gone with DropBox simply because so many more apps can get at those files. If I had very sensitive data to store, I could use ownCloud at the cost of less ease of access.
\nOn OS X I am huge fan of 1Password. However, that’s not going to help me on Fedora. The best solution that I could find was LastPass, which runs as a browser plugin under OS X, Linux, and anything else that can run modern builds of FireFox, Chrome or Safari.
\nLastPass is free for basic services, but if you want a native mobile app, you need to subscribe. My cheapskate workaround is simply to log into LastPass with my account and manually copy passwords when I need to use them on iOS.
\nCloud services offer a great way to level the playing field between Fedora 20 and OS X. By shifting my cloud usage to apps that could leverage DropBox or presented well-know web services APIs, I was able to create a setup that makes Fedora 20 work as well as OS X and still be able to get at the same data from my iOS-based mobile devices.
\n\n DropBox shows up a lot in this article. The main reason is that DropBox offers ubiquitous file storage with widely adopted support. Many iOS apps continue to offer DropBox-based syncing as an alternative to iCloud, and DropBox runs on Linux systems as well. This has inadvertantly placed DropBox as the best low-level cloud storage option for a wide variety of uses. ↩︎\n
\nA little background here: I have been a Mac enthusiast for a number of years now. I would not say that I am religious about OS X, but if the OS X user experience is a philosophical application of “opinionated software”, then I find myself in agreement with most of the opinions that the Apple UX team has expressed.
\nHowever, I am also an open source developer. I’d like to believe that it is possible to create a similar and possibly even superior experience with a Linux-based desktop environment. And lo, this is what lead me, a few weeks ago, to get Fedora 20 running on a MacBook Air. If you’re interested in trying the same thing, check out Matt Hicks’ invaluable blog post on setting things up.
\nOver a series of blog posts I am digging into my impressions of the Fedora 20 user experience as I work through this total switch-over.
\nI spend 95% of my computer time in front of the same 5 apps:
\nI typically also have several helper apps that I call upon and I usually have music going in the background, but I cover “Cloud stuff” in the next part of this series.
\nFor my text editor and web browser, I’m using the same tools on both platforms: vim for text and Firefox for web. The experience of these between OS X and Fedora 20 is 99% identical. Not much else to say there.
\nNow as you read on you may be thinking: “hey, you could have identical experiences all the way through if you just used <open_source_tool> in place of <os_x_native_tool>”. And you’d be right. But in the cases below, I really feel like the OS X native tool is a little better, and in each case I will attempt to explain why.
I don’t always think that the right answer is to make Linux match OS X, but when OS X does something really well, I wish the open-source equivalent could do the same. In the case of Thunderbird, the functionality that I desire is frustratingly close, but not quite there. Specifically, I am talking about the way that Mail.app handles threaded e-mail discussions, both in the e-mails list and the email display field.
\nTo Thunderbird’s credit, you can edit the columns displayed in the e-mails list, turn on threaded views, and Thuderbird even kinda sorta shows an all-discussion view of the thread. But Mail.app’s multi-line mail list entries make for a narrower emails listing field, and the all-discussion view still shows complete, distinct e-mails from the thread – including your own e-mails which may have gone to the Sent folder but still count as part of the conversation1. Thunderbird doesn’t do this even though someone first raised this issue in 2003 and people have periodically asked for it ever since.
\nBut while the threaded views leave something to be desired, I will be the first to say that Thunderbird’s e-mail search is awesome. It is very fast, includes a cool chart of word hits by year, and shows enough of each e-mail in the search result to give better context for the word being searched.
\nThunderbird isn’t the only mail client for Fedora; in fact, it isn’t even the default client for Fedora. But while Evolution offers a very Outlook-like experience, the Outlook that it emulates is Outlook 2003. I don’t want to get into a colors and buttons argumnt, but Evolution looks pretty dated and has similar problems to Thunderbird with regard to offering an information-dense layout. Moreover:
\nEvolution lacks the sort of plugin ecosystem that Thunderbird has. These plugins add a ton of value and help Thunderbird to stay current as new kinds of web services arise.
\nEvolution doesn’t allow you to map mail boxes to known roles, like for instance having an Archive folder. So if you aren’t using Gmail and you want Archive-like behavior, Evolution isn’t for you.
\nI’d love to be proven wrong, but as far as I can tell, nothing in the open-source space can even come close to what Apple, Google and Microsoft have done with their calendaring apps. I understand that doing calendaring correctly is a really hard problem, but the net result is that on the client side, open source systems seem to be limited to only the most basic functionality.
\nIf you only keep calendars for yourself, this is probably not an issue. But for anyone who wants to schedule a meeting or have access to delegate calendars, you are going to run into problems. For this reason, and because I’d already chosen Thunderbird over Evolution, it seemed logical to pick Lightning for e-mail. Lightning is “good enough” for showing me my schedule at a glance, but doesn’t distinguish itself as being a more functional calendar app than any of the other Linux options (like Evolution, or even the nascent California).
\nI use ZNC as a proxy server for several IRC servers where I hang out on a daily basis. ZNC’s primary job is to keep a log of the discussion on the channels that I follow so that if I miss something, I can read about it and follow up later. In order for my ZNC server to know which real IRC server I want to connect to, I need use a log name of the form:
\n<znc_username>/<znc_config_name>\nWell, you’d be surprised to find out how many IRC clients can’t tolerate a / in your username. On OS X, this drove me right into the arms of LimeChat. I’m really glad it did, too, because the information density jammed into LimeChat’s subdued interface is totally amazing.
Many IRC clients on Fedora 20 suffer the same username handling problem, but HexChat stands out for being able to connect to VNC. Additionally it can be configured to look almost identical to LimeChat, save for LimeChat’s awesome extra window pane that shows you a continuous stream of all of the convos from the IRC room you aren’t currently looking at.
\nFrom my perspective, Thunderbird, Lightning and HexChat provide a workable experience on Fedora 20. The more time I spend using them, the more I can work around or otherwise just accept those little bits of polish that are lacking compared to OS X. In a few cases, as with Thunderbird’s search, I can even see myself missing that functionality when I’m in front of the OS X native app equivalent. But there’s room for improvement on the Linux side and I’d love to see some open source projects that can close the gap.
\nNext up in this series, I’ll be talking aout the cloud services that I use and how I modifed things to use them from both OS X and this Fedora 20 system.
", "date_published": "2014-08-18T22:04:07+02:00", "url": "https://harrisonripps.com/2014/08/18/os-x-to-fedora-the.html", "tags": ["Tech and Work"] }, { "id": "http://djharrisonripps.micro.blog/2009/03/16/protecting-important-data.html", "title": "Protecting Important Data", "content_html": "My recipe for not losing important stuff on my home desktop system…
\nI set up a pair of disks in a RAID 1 configuration to be the “home” folder on my system. RAID arrays can be accomplished numerous ways. I do this in software under Linux; under Windows you are more likely to end up using a semi-hardware solution. However, disk redundancy only protects you from a single disk failure. If you nuke a file on your RAID pair, it’s still gone in both places. That brings me to:
\nrsnapshot is “Time Machine for the rest of us”. Ive got it set up to push hourly incremental backups out to my external USB drive. On Windows, I had great success with Acronis True Image Home in differential mode, but Norton Ghost is an old standby. Next step:
\nDoing what I do protects me against most data-loss woes, but it won’t protect me against the meteorite that slams into my home office. For that, theres the concrete bunker Im working on in my spare time and off-site backups. While there are solutions that use cross-network syncing for the ultimate in worry-free backups (unless you worry about data security), I’d rather put my faith in “sneaker net” and a second external hard drive. Swapping external hard drives on a weekly basis and carrying the “off drive” to the office would mean never losing more than a weeks worth of important stuff…
\nWhat do you do to protect your important data?
", "date_published": "2009-03-16T17:45:00+02:00", "url": "https://harrisonripps.com/2009/03/16/protecting-important-data.html", "tags": ["Tech and Work"] }, { "id": "http://djharrisonripps.micro.blog/2008/12/19/how-not-to-optimize-a.html", "title": "How not to optimize a site with AJAX", "content_html": "I’ve inherited a project that is seeing some performance issues. We knew this was at least partly to blame on the UI, but most of the problem is down in the application. Fine; we’ve got a plan for the application-level issue and that’ll get solved. But right now, I’ve been picking through the UI because even though inspection with FireBug reveals that the site is loading at a reasonable clip, it doesnt seem to be rendering the main page until every last linked resource is loaded in the browser.
\nAs the subject of this post should indicate, this project features AJAX. AJAX isn’t exactly brand new, but this project represents the first aggressive use of AJAX in my particular custom-projects corner of the universe. And as far as I can tell, once the decision was made to “go AJAX!”, a side directive of “abandon sanity!” was also put in motion.
\nHeres what’s great about AJAX: You load the static elements of your page once, and then you refresh the dynamic stuff with a behind-the-scenes request that leaves the static page elements unchanged. In other words, it makes a web page behave more like a locally installed application. But let’s say you’ve just learned about AJAX and you want to GO the DISTANCE! Then, even though the static page elements should never change, maybe you dynamically request them, too! And then use JavaScript to discretely rewrite the whole page! In your zero-latency development environment, the difference in rendering time may be completely undetectable. Unfortunately, when half a planet’s worth of internet lies between you and the server, the effect is pretty different.
\nYour AJAX request is an extra layer of abstraction that just linking static elements from the page itself will completely avoid. And, because the associated rendering may not be possible until after you’ve finished loading all of the AJAX-related JavaScript resources, you leave your users staring at the bare rafters of your web site until the loading is done.
\nMoral of the story: identify static page elements and keep them out of your AJAX scheme. If you really want to do the whole thing in AJAX, at least give the user a meaningful static site to look at while all of the background requests are churning. They might stay on your site long enough to see what you’ve cooked up..
\n", "date_published": "2008-12-19T04:00:00+02:00", "url": "https://harrisonripps.com/2008/12/19/how-not-to-optimize-a.html", "tags": ["Tech and Work"] }, { "id": "http://djharrisonripps.micro.blog/2007/07/19/sunrocketd.html", "title": "Sunrocket'd", "content_html": "Well, that was a bad call. SunRocket basically imploded this week, and I’m unlikely to see a prorated return on my flat $200 for a year of service. So now I’m on the fence; I’d sure hate to come crawling back to Vonage in time to watch them implode the same way.
\nSunRocket’s “sorry we screwed the pooch” e-mail provides links to Packet8 and TeleBlend, who are offering deals to the 200K folks that are now looking for a new phone service. Well, I’ve heard of Packet8, but why does the name “Teleblend” give me an image of someone running a VoIP server and a media gateway in their basement? Personally, I think my experiment with VoIP may go on hold for a while. It may be time for a Triple Play…
", "date_published": "2007-07-19T18:00:00+02:00", "url": "https://harrisonripps.com/2007/07/19/sunrocketd.html", "tags": ["Tech and Work"] }, { "id": "http://djharrisonripps.micro.blog/2007/05/24/farewell-brix.html", "title": "Farewell, Brix", "content_html": "After giving the whole thing some serious consideration, I am ending my career with Brix Networks. I have had a tremendous seven years there, and will miss working along side some of the smartest engineers I could ever hope to meet; chief among them, my long-time friend and mentor Mr. Bennett. I have accepted the position of Director of Technology at Mark Altman & Associates, a small and successful publishing agency that is looking to add technical leadership as part of their plan for business growth. Many thanks to K. for bringing this opportunity to my attention. Most importantly, thanks to my wife for supporting me in this decision–even as we prepare to bring kid #2 (!) into the world. Obvious symbolisms aside, this is a great opportunity for me and I look forward to the new challenge.
\n", "date_published": "2007-05-24T18:00:00+02:00", "url": "https://harrisonripps.com/2007/05/24/farewell-brix.html", "tags": ["Tech and Work"] }, { "id": "http://djharrisonripps.micro.blog/2006/10/16/when-does-a-resume-become.html", "title": "When does a resume become a C.V.?", "content_html": "When you publish something! And let me tell you, its a page-turner. Oh well; you gotta start somewhere :-)
\n", "date_published": "2006-10-16T18:00:00+02:00", "url": "https://harrisonripps.com/2006/10/16/when-does-a-resume-become.html", "tags": ["Tech and Work"] } ] }