|
jiggerypokery posted:I am applying for CTO/Dev team lead roles (if I apply at all) so the expectations might be a little different than for junior/senior Dev roles. I can talk about deployment process and architecture all day long, unfortunately for anyone in the vicinity. Basically it's come up that I need to show that I can architect software with a modern stack at a lower level so that it's clear I can perform code reviews and mentorship. I'm comfortable with CI/Deployment/infrastructure. The issue is that I feel a little out of touch specifically with the node/typescript ecosystem. At your experience level I believe you should have no issue getting yourself up to speed on a ts/node portfolio project
|
#
?
Oct 19, 2020 13:11
|
|
|
|
| # ? Jun 5, 2024 05:56 |
|
|
Tea Bone posted:My company is moving from a third party host to our own hosting. quote:What I'm concerned over is: quote:B) When we switch over to our new hosting, would the new certificate cause any confusion for browsers expecting the old one?
|
|
#
?
Oct 19, 2020 15:59
|
|
|
Anyone used any website diff tools? We're migrating from a clunky homebrew CMS to a WordPress backend over the next few months. Same content, same design. The project lead said we need to manually track every change we make to live content between now and then in Excel so we're sure we make the same changes in WordPress. That seems like a pain in the rear end and some little but important change is going to be forgotten to be added to Excel. I'm considering writing a scraper in Python that can compare the stuff that would matter, but I know there are tools out there that do this already.
|
|
#
?
Oct 21, 2020 19:32
|
|
|
are you trying to diff the actual pages or diff the content? modern wordpress has a full API for everything content-wise (returns slug, title, body, fullbody if read more, etc) otherwise something like just GETing both versions and removing all tabs, newlines, spaces / using a html minifier library and something to convert into formal html (like normalise <img /> into <img>, etc) would probably work
|
|
#
?
Oct 21, 2020 19:49
|
|
|
How are you migrating things... by hand? Normally for my [whatever platform] to WordPress migrations I use a combination of custom scripts and import plugins (notably WP All Import) to define a workflow for performing an import from a given export file (XML, CSV or other). Once the workflow is set up properly I can use it any number of times to perform imports. So generally I'll do something like this:
|
|
#
?
Oct 21, 2020 20:04
|
|
|
On the actual migration process, I have no idea what unholy rituals, exports or manual work is being done to move our stuff into WordPress. A different team is doing that work. I shudder every time I have to access the homebrew CMS and make changes to content. I haven't seen the homebrew database schema (if any), but there are about 50 "fields" that can be used on a page and are used inconsistently for anything from a section heading to text on a button, depending on the page. So I can't imagine any useful export or mapping. What I'd like to do is avoid the time-consuming and error-prone manual tracking of changing "the" to "that" in Excel. I'm seeing use of either a visual diff tool (which would also catch CSS mistakes on color/placement) or scrape-and-minimize comparisons of the important HTML bits. Just curious if anyone's ever used any of the available tools for that. Scraping I can do myself if that's the only or best option. It might be a more appropriate question in a web app development thread, similar to automated cross-browser testing.
|
|
#
?
Oct 22, 2020 00:13
|
|
|
Ooof. I wish I had a good tool to share with you, because your migration process sounds terrible. Good luck with it!
|
|
#
?
Oct 22, 2020 03:12
|
|
|
My page speed score obsessed company is looking at GTMetrix and their YSlow score. Under "Use a Content Delivery Network (CDN)" it suggests that I load a couple of things using a CDN. Mostly "above the fold" images such as the hero image that isn't lazy-loaded, and the site's main CSS file. I don't think using a CDN for around 300kb of images and CSS will speed up the site at all and instead just add more HTTP requests, but I wanted to ask just in case. The sites aren't slow at all, but the YSlow score is "only" at 75%.
|
|
#
?
Oct 27, 2020 18:43
|
|
|
LifeLynx posted:My page speed score obsessed company is looking at GTMetrix and their YSlow score. Under "Use a Content Delivery Network (CDN)" it suggests that I load a couple of things using a CDN. Mostly "above the fold" images such as the hero image that isn't lazy-loaded, and the site's main CSS file. I don't think using a CDN for around 300kb of images and CSS will speed up the site at all and instead just add more HTTP requests, but I wanted to ask just in case. The sites aren't slow at all, but the YSlow score is "only" at 75%. For landing pages, I always look at GTmetrix scores in comparison to the top 5 pages for my top 2-3 keywords. If I’m as good as the best, I don’t care. I think comparing them might help your customer feel they’re doing great and could invest better elsewhere. That said, use the inspector to see what’s killing you. I got a massive speed up and gtmetrix bump with 60 min of work by serving static gzipped via cloud front but this was serving ~2MB of text/js/css static.
|
|
#
?
Oct 27, 2020 18:47
|
|
|
CarForumPoster posted:For landing pages, I always look at GTmetrix scores in comparison to the top 5 pages for my top 2-3 keywords. If I’m as good as the best, I don’t care. Is there a way to do that in GTMetrix Pro for example, or do you need to click Compare and compare each site manually? This is a Wordpress site. A couple of things it says to serve via CDN are external scripts, such as tracking pixels and Google Analytics. The site initially loads 1.17MB of content. It's on Media Temple, which I know isn't the fastest host out there, but I can't convince them to use anything better, and they don't have anyone who can manage something like a VPS. Maybe SiteGround would be the best?
|
|
#
?
Oct 27, 2020 19:05
|
|
|
What are you doing to have 300 KB of CSS and one image? Giant header in png? You can ignore those mentions, they are likely already served via CDN. Your analytics, at least the ones at the bottom (not inlined RUM / etc) should be defer/async anyway
|
|
#
?
Oct 27, 2020 19:07
|
|
|
Biowarfare posted:What are you doing to have 300 KB of CSS and one image? Giant header in png? It's not one image, it's a slider. It's 162 KB on mobile because I'm serving smaller images, but I can't compress these images any more than they are for desktop. I hate sliders, I think they're useless and most people aren't going to stick around to watch a few seconds of images of houses slide by.
|
|
#
?
Oct 27, 2020 19:20
|
|
|
LifeLynx posted:It's not one image, it's a slider. It's 162 KB on mobile because I'm serving smaller images, but I can't compress these images any more than they are for desktop. I hate sliders, I think they're useless and most people aren't going to stick around to watch a few seconds of images of houses slide by. Sounds like the perfect use case for lazy loading. Also ironic that the platform with higher resolution screens gets the lower resolution images.
|
|
#
?
Oct 27, 2020 20:06
|
|
|
MrMoo posted:Sounds like the perfect use case for lazy loading. Also ironic that the platform with higher resolution screens gets the lower resolution images. You're right, I lazy-loaded them and it's much better.
|
|
#
?
Oct 27, 2020 20:53
|
|
|
I have divorced myself from a Project From Hell and it's pretty much the best feeling I've ever had other than busting a nut. It was a long time coming and it was a difficult decision but I can tell you the other side of it is glorious, and as much as I'd wished I heard this from other people more often, it's now my time to let other people know.
|
|
#
?
Oct 29, 2020 04:03
|
|
|
Quitting stuff is the greatest feeling in the world.
|
|
#
?
Oct 29, 2020 05:32
|
|
|
prom candy posted:Quitting stuff is the greatest feeling in the world. It's this. I should've quit when I first thought about quitting, instead of dragging my rear end to the point where it was making me miserable, and then quitting anyway and ultimately ending up in the same place. Hear ye, all those who read my shitposts, and learn from my mistakes.
|
|
#
?
Oct 29, 2020 17:47
|
|
|
Hey, any recommendations for a simple+free analytics I can integrate that isn't overwrought if I just want simple, like, visit count and maybe a map, that'd be quick and fast and not affect times userside and doesn't go and shove its tentacles all over the page?
|
|
#
?
Nov 11, 2020 12:51
|
|
|
MikeJF posted:Hey, any recommendations for a simple+free analytics I can integrate that isn't overwrought if I just want simple, like, visit count and maybe a map, that'd be quick and fast and not affect times userside and doesn't go and shove its tentacles all over the page? maybe something here? https://github.com/onurakpolat/awesome-analytics#general-analytics
|
|
#
?
Nov 11, 2020 13:28
|
|
|
Inacio posted:maybe something here? https://github.com/onurakpolat/awesome-analytics#general-analytics Oh, thanks. I'm thinking... GoatCounter?
|
|
#
?
Nov 11, 2020 13:45
|
|
|
MikeJF posted:Hey, any recommendations for a simple+free analytics I can integrate that isn't overwrought if I just want simple, like, visit count and maybe a map, that'd be quick and fast and not affect times userside and doesn't go and shove its tentacles all over the page? I really like PostHog because all the data is stored on your server so its first party analytics not blocked by adblockers. Free, super easy to install and you deploy the code to heroku with one click. EDIT: Also my fav thing about it is the clack/teams notifications I can set for certain actions. E.g. when someone clicks the button that indicates a conversion, we get a message to slack or MS teams. Setting this up took about 5 minutes and is super useful for onboarding. CarForumPoster fucked around with this message at 15:43 on Nov 11, 2020 |
|
#
?
Nov 11, 2020 14:58
|
|
|
Does anyone actually use JPEG 2000? I was quite surprised to stumble upon a webstore that only uses JPEG 2000's, but hiding as regular JPEGs, https://www.inthestyle.ie
|
|
#
?
Nov 12, 2020 22:20
|
|
|
MrMoo posted:Does anyone actually use JPEG 2000? I was quite surprised to stumble upon a webstore that only uses JPEG 2000's, but hiding as regular JPEGs, A lot of medical imaging modalities have been using JPEG2000 for storing dicom images (http://dicom.nema.org/medical/dicom/2016c/output/chtml/part05/sect_8.2.4.html). JPEG2000 has the advantage of supporting real "lossless" compression as well as lossy. I don't see JPEG2000 used much outside of medical imaging anymore.
|
|
#
?
Nov 12, 2020 22:32
|
|
|
Hey bros. Looking for feedback on a website. I launched it a few months ago, and am getting ready to launch a new, relatively-high-volume product. I'm proficient at making web apps etc, but am no good at making websites pretty or modern. Link My thought (tell me if I'm wrong!) was since the products are specialized, I can get away with an austere website as long as it's clear. It's mostly server-rendered, with no JS dependencies, and a small amount of DOM-manipulation JS to manage the shopping cart and checkout. (Might change it to Typescript or WASM to improve code quality, but it won't make a meaningful difference, and I enjoy not having a build step) Are there UI problems? What's the #1 change you would make? At the risk of getting political, I'm fed up with the state of modern websites, both internal (JS frameworks and dependencies that negate advancements in internet and computer speed), and external (tacky marketing like trusted by, cartoon people pointing at graphs), but don't want to hurt myself for controversial/political opinions. Thank you. Dominoes fucked around with this message at 02:24 on Nov 16, 2020 |
|
#
?
Nov 16, 2020 02:09
|
|
|
Alot of the pages have boxes and images that aren't reaponsi ve on my device 
|
|
#
?
Nov 16, 2020 02:12
|
|
|
Oh gently caress. That's exactly what I needed to see. Diving in. What's your device? The responsiveness code is pretty naive; sets a screen width threshhold. Are you zoomed in? edit: I can rep the pH module thing. Fix inbound. Dominoes fucked around with this message at 02:17 on Nov 16, 2020 |
|
#
?
Nov 16, 2020 02:13
|
|
|
Not zoomed in, I'm on a Samsung 10se
|
|
#
?
Nov 16, 2020 02:22
|
|
|
Thank you. Does the pH module page look any better now? Added a flex-wrap to the buttons. The image might still be bugged.
|
|
#
?
Nov 16, 2020 02:22
|
|
|
Yep, front page is still odd but the pH page is good
|
|
#
?
Nov 16, 2020 02:24
|
|
|
Awesome! I'd hoped there would be a space between the buttons, but it may be a CSS file caching thing. Thank you so much, btw.
|
|
#
?
Nov 16, 2020 02:25
|
|
|
Is the index any better now? Just did some div-wrapping to try to mimic the other entries better.
|
|
#
?
Nov 16, 2020 02:32
|
|
|
This site looks bad and is a waste of your considerable talent and time. Why not just make a Shopify website or literally any site that trivially lets you cross list your products in Google Shopping, Amazon, eBay, etc.? It seems like rolling your own here provides no value to you or the users. My site feedback is that it looks poorly optimized for SEO and cheaply designed such that I wouldn’t trust giving it a credit card. I think you’re putting in lots of effort for a much worse site and it’ll directly hurt product adoption. Make it as easy to buy as possible. Accept PayPal. There are lots of good times to roll your own. This isn’t one of them. Use a site builder if all you want to do is make an e-commerce site.
|
|
#
?
Nov 16, 2020 02:38
|
|
|
Yeah it's 2020, don't make your own website unless you are or want to be a web developer
|
|
#
?
Nov 16, 2020 02:41
|
|
|
Thank you. Do you have a specific service you'd recommend? How do Shopify etc compare to SquareSpace etc? I think one of the reasons I rolled my own was to maximize speed, not deal with trackers etc, but perhaps the juice isn't worth the squeeze. I'm willing to start again and toss the existing site if it's the right answer. There's no app-like UI that would require a custom site. What would you change re SEO? That's also an area I have no idea how to do. I think I have a slight edge from backlinks from related sites (Github, and various language-specific package managers). Searching anyleaf on all search engines pulls up a first page that looks right. ph module hits midway through the first page on DuckDuckGo, and top of second on Google. ph raspberry pi and orp raspberry pi hit first page of DuckDuckGo, but nothing on Google. I would love to improve results, since I think search is what drives most of the traffic. Getting search terms like the latter few (and related) on the first page of Google would help a lot, I think. Worth considering; I may be completely wrong here, but I think I can get away with a more bare-bones, functional website due to the audience, in the vein of Hacker News. Or maybe that's killing sales and SEO. Dominoes fucked around with this message at 03:17 on Nov 16, 2020 |
|
#
?
Nov 16, 2020 02:57
|
|
|
Dominoes posted:Thank you. Do you have a specific service you'd recommend? How do Shopify etc compare to SquareSpace etc? I think one of the reasons I rolled my own was to maximize speed, not deal with trackers etc, but perhaps the juice isn't worth the squeeze. I'm willing to start again and toss the existing site if it's the right answer. You don't have enough content on your potential landing pages. They're note keyword optimized. You don't have images with appropriate text and alt tags to rank in image search...or at all. There's basically no SEO here. There are lots of good primers on SEO training and how to do keyword research. I'd put the clickminded SEO course up there but it is spendy. I'd only worry about Google and maaaaaybe Bing. You don't need to think about DuckDuckGo. Sadly, I dont have a good ecommerce suggestion except for not Weebly. It's been so long since I ran an ecommerce site (nearly 10 years!!!) that I dont think I could offer great advice. That said, Shopify seems to be what everyone talks about. I'd come up with requirements for the site and find a site builder that meets the most important ones. Ten years ago, I found that easy cross listing, even if only by a plugin or Excel export, and easy setting up of landing pages/SEO optimization were valuable requirements. Do keyword research for your products. Get your products on Amazon/eBay with the appropriate markup for their fees. Its very cheap advertising. I highly recommend ahrefs. Spendy-ish at $100/mo but you can cancel anytime. I've cancelled for a few months and reactivated twice now. I'm an aquaponics "maybe one day"/dilettante who has been "meaning to do a project" and has a box full of arduinos/circuit python boards/various sensors/drivers/steppers from previous tinkering things. I'm probably pretty close to your customer. Hit me up when you make some progress and I'll happily dive in more to help you stand out between you and competing products. Sites an absolute disaster though and I say this harshly only to encourage you to focus at what you're great at knowing the quality posts you make in the Python, Electronics, etc. threads. Learn the SEO fundamentals, get your products listed on other sites.
|
|
#
?
Nov 16, 2020 03:17
|
|
|
Thank you for the details. These things are tough to identify first-hand (especially anything related to aesthetics), so your feedback is very helpful. Going to do bandaid fixes like add alt tags to the images ASAP. I missed out on that entirely: The only way any relevant images show on Google image search is if I search anyleaf, which is pointless. Another great point on identifying keywords. I've been guessing (eg ph raspberry pi, ph arduino etc), but haven't validated this, and it's subjective and... I'm probably guessing wrong. I can't get on Amazon; they won't verify my utility bill, after 20 appeals, and support is useless. :/ Didn't consider eBay since I associate it with used products, but that may be an error. I would love to get on Amazon, but have given up for now. Had bad results from Google Ads. (Ie advertising cost more than profit), but that's worth trying again after fixing/replacing the website. Caveat: There is a potent school of thought, that says don't pay anything for marketing in the beginning. This assumes you're reaching out though, and I've put that on the backburner while working on the upcoming device. I'm def interested in feedback on relations to competition. My (Pidgeonholed) perspective is it's a middle-ground between poorly-documented Chinese parts, and western ones pricier than the audience has an appetite for. Dominoes fucked around with this message at 03:41 on Nov 16, 2020 |
|
#
?
Nov 16, 2020 03:35
|
|
|
Dominoes I really really admire your positive attitude. That’s so hard to come by and your openness to critique and change is awesome. That’s a quality I really respect and try to do better with.
|
|
#
?
Nov 16, 2020 04:15
|
|
|
Dominoes posted:Thank you for the details. These things are tough to identify first-hand (especially anything related to aesthetics), so your feedback is very helpful. Going to do bandaid fixes like add alt tags to the images ASAP. I missed out on that entirely: The only way any relevant images show on Google image search is if I search anyleaf, which is pointless. Why not make a thread for this venture in BFC? It seems like you’re at that stage now. You’re just launching and it seems like you could use some business and marketing advice. Things like: -A billion Chinese spammers make it on Amazon, so can you! -I agree, your marketing budget at this venture should be very low but your marketing effort should be very high! SEM and paid search is where money goes to die. -What about getting distribution through adafruit, sparkfun and the top 10 search results for things like: DIY aquaponics, aquaponics parts, water quality parts, (other applications), etc. -middle of the road products are often extremely risky as a real business
|
|
#
?
Nov 16, 2020 04:40
|
|
|
Thanks for the feedback Violator. There's no benefit to obstinance when people are going their out of their way to be helpful. Unaddorned critique is the most helpful. I'll make that BFC thread once I finish the Water Monitor, since that will broaden the audience, and change the business dynamics considerably. I completely agree on getting it on Sparkfun. From what I can tell, Adafruit lets you send your design; they'll make and sell it, then give you a portion of the sale. I sent them an email. Sparkfun lets you do that, or sell directly: Info. I'll try ebay, and call Amazon again today. While the current products are middle-of-the-road (A niche is another way to look at it), the upcoming one doesn't have a feature-matched competitor, which should help. Dominoes fucked around with this message at 20:55 on Nov 16, 2020 |
|
#
?
Nov 16, 2020 17:39
|
|
|
|
| # ? Jun 5, 2024 05:56 |
|
|
I'm (un/fortunately) getting a lot of content migration work these days. I've already completed a couple of projects that required migrating tens of thousands of records from site A to site B, and while they went fairly well, I learned I need to be more specific when setting expectations with clients. I'm curious if any of you have done any big migrations, and if so how you talk to clients about testing/ensuring content fidelity? I'm trying to build some language into my contracts that basically says, "look, I'll run automated tests on your content and will manually review X% of the items we migrate, but I'm not going to look at every single one of your ten thousand posts to make sure there are no minor errors the tests missed." Does anyone have opinions or experience in this matter? Thankfully my previous clients have been pretty chill about this and I've been able to correct most of the issues that have popped up, but moving content between completely different platforms is a messy business, and it seems inevitable that some image is going to go missing, or a quote is going to go unescaped, or what have you. My next project involves content that has already been migrated at least once before this migration, so it's already going to be even messier. I want to make sure I cover my rear end. kedo fucked around with this message at 17:57 on Nov 19, 2020 |
|
#
?
Nov 19, 2020 17:52
|
|