G

Most UK businesses I audit are burning serious money on digital marketing… with zero real ROI. I fix that. I’m Graeme, founder of Strathmark Consulting. - Independent digital audits - High conversion websites & web design - Technical fixes (Core Web Vitals, speed, conversions) Strategy that actually drives revenue, not clicks! Proof: Through my subsidiary apexaesthetics.co.uk, one aesthetic clinic went from 12 to 120+ patient leads per month (+900%) after I rebuilt their site + local SEO. I run everything while travelling full time. Need an honest review of your digital setup - no nonsense audits and never any minimum term contracts)? DM me or head to

@knz_dev @joshwoodward What's wrong with the DMG?

@joshwoodward Does anyone already have the Homebrew Formulae?

Introducing Gemini on Mac. We heard your feedback. We recruited a small team. They built 100+ features in less than 100 days. 🤯 100% native Swift. Lightning fast. Let us know what you think!

@JesusGodAndKing @NoahHirshon @joshwoodward What's this about adding max effort in terminal? Is this still applicable?

@NoahHirshon @joshwoodward Claude Code: x.com/JesusGodAndKin…

@oggbarcelos @joshwoodward Exactly. ChatGPT has had this since day. It's so tedious doing it the long way.

@joshwoodward Please: support paste images inside prompt input. 🙏 It's not allowing it yet.

@dang5_them @imCubiek @joshwoodward That's what I was thinking. What am I missing?

@imCubiek @joshwoodward Doesn’t iOS already have native Gemini app ?

@meetjamie_ai 3-4 minutes if I've setup the meeting. 60 seconds if they've set it up.

Meeting Etiquette: How long do you wait before ending a call when the other person doesn’t join?

@meetjamie_ai Used your product for the first time today. To say I was impressed would be an understatement. Flawless! Even with Scottish and Irish accents.

Finally! Looking forward to this. Now can you please add the ability for it to produce XLSX, PDF and give it a proper working memory like ChatGPT. Projects are also essential. I've switched from ChatGPT to Gemini for 70% of work. But it's still seriously falling behind in those departments to GPT. It's tedious to explain my background, task and project every single time. ChatGPT knows what I'm working on and what I do. Geminis memory is all over the place.

This is definitely the way to do high level mission critical work when you've got the budget. Nice explanation. Google's Jitro is due out soon and I look forward to emulating the same flow. It really is crazy how you can now replace a team of Devs overnight and have a better end result. Boggles the mind.

I work in a large monorepo that includes multiple front ends, two API servers, and a native iOS app. It’s the backbone of the company I work for, and includes administrative tooling ranging from team management, client management, sales pipelines, manufacturing resource planning, and more. Most of the time is spent on review gates. My main session acts as an orchestrator, hires feature and interface planners and expert panels of consultants for creation and review of plans, and then implementation agents. The first hour is generally used for meta planning. Teams and panels work to review, create, and refine shaping docs, feature specs, and UX goals and usage patterns. These are then converted into a comprehensive series of orchestration and implementation plans, broken down into waves and sprints. After each wave of implementation, a series of 9+ specialized agents review the implementation for things low code quality, bugs, regulatory compliance, user experience and usage flow, and any drift from initial specs. Additional agents also implement tests, dogfood with browser use, and at least two adversarial agents try to break the work done in said wave. Then, a team takes into account all findings, drafts intermittent plans to address any issues found, and then new wave of implementation agents go in to correct any issues. This review gate is repeated ad nauseum until two consecutive review cycles find no medium+ priority issues. (Low priority issues are often not with the tokens to fix and are just unnecessary nitpicks) Then proceed with the next wave with the same process. After all waves are complete, several more review cycles are run until the initial requirements are fully met. It can be very token intensive, but the results are wild and better than anything our team of two engineers could produce on our own (we are a small company)

We've redesigned Claude Code on desktop. You can now run multiple Claude sessions side by side from one window, with a new sidebar to manage them all.

@DeryaTR_ I don't see anyone complaining about this. It is incredibly good value. People rightfully complain about Claude Codes limits, but never ChatGPT. Can anyone show me actual complaints or moaning posts on this subscription? Or is this just a strawman for engagement?

For $20 a month, you can create multiple websites or apps using Codex AI coding that would have cost 100x that! This is not even considering the immense benefit of ChatGPT thinking. Yet people complain about a cost that’s less per day than a crappy small coffee at a gas station!

@Windywowwow @rappo01 @glen_mountain @TheGriftReport 😂

@rappo01 @glen_mountain @TheGriftReport It’s Liverpool, mate.

39-year-old grandad-to-be Andrew Dickson from Blackpool was caught at Liverpool John Lennon Airport with 25.55kg of cocaine worth £1.9 million. He arrived on an easyJet flight from Alicante with the drugs hidden in 22 blocks inside his suitcase, purity around 80%. Dickson claimed it was “holiday gifts” from a friend and said the suitcase wasn’t even his. He later admitted acting as a drugs mule to clear a debt from his long-term cocaine and alcohol addiction. Liverpool Crown Court sentenced him to six years and eight months in prison. Sickening.

Kezia Noble chooses Reform UK over Rupert Lowe’s Restore Britain primarily because she believes Reform offers not just shared values and passion (“heart”), but a detailed, practical, implementable plan plus the real-world offline organisation needed to actually win elections and deliver change. She has explained this repeatedly on her own show UK Now, in YouTube videos/debates, and in direct X posts — including in the very thread you linked.Here’s the in-depth breakdown straight from her own words and statements:1. Shared Values + “Heart” (She Acknowledges This Upfront)Noble openly agrees with many of Restore Britain’s core positions. In her viral post that started the exchange you linked, she listed the exact priorities she (and most right-wing voters) support:Brexit Opposition to lockdowns Low taxes Extremely limited immigration Supporting small businesses Being tough on crime Building a better future for children She has repeatedly said Restore Britain has “a lot of heart” and genuine passion. She is not dismissive of them — she even called Reform’s decision to suspend Rupert Lowe a mistake in some earlier comments. The overlap in goals is real, which is why the question “why not join Restore?” keeps coming up.2. The Deciding Factor: Reform Has a Workable, Implementable Plan (Not Just Slogans)This is the #1 reason she gives. In a widely shared video post from 16 March 2026 (captioned “For me it’s like this…”), she said:“Restore Britain does have a lot of heart. (Not saying that in a condescending manner) But Reform U.K. has a lot of heart and a workable plan that can ACTUALLY be implemented. Plus very strong OFFLINE engagement. You need both to win.” She stresses: “It’s not a plan if it can’t be implemented.” Reform publishes detailed, point-by-point policies (she has done entire videos breaking down their immigration plan line-by-line, for example). Restore Britain, in her view, is newer, more online-focused, and lacks the same level of concrete, ready-to-execute detail and infrastructure.She has dedicated full episodes and debates on UK Now and her YouTube channel to this exact comparison:“Who Actually Has a Plan? Reform UK vs Restore Britain” “Reform UK vs Restore Britain: TIME FOR A REALITY CHECK” In those, she walks through policy depth, credibility, and feasibility.3. Offline Organisation and Electoral InfrastructureNoble repeatedly highlights that online buzz alone isn’t enough. Reform has:A large existing candidate network (she is the Reform candidate for Kensington & Chelsea and has been out canvassing/leafleting with dozens of others) Established local presence and ground game National visibility and momentum Restore Britain is strong online (especially on X) but, according to her analysis, lacks the same offline structure and proven electoral machinery right now. She believes you need both heart and infrastructure to break the “uniparty” and actually win seats.4. Pragmatism and “Real Chance” of Fixing BritainWhen she joined Reform (she had never been a member of any party before), she stated publicly:“Reform U.K. are the only party that’s got a real chance of fixing this country. Time to unite.” She sees Reform as the vehicle with:Established credibility Broader appeal beyond the “X bubble” High-profile leadership and track record Enough scale to avoid splitting the vote (a risk she associates with newer parties) She has debated pro-Restore guests on her show and still concluded that, outside of social media, Reform is the more realistic option right now.5. Minor Policy NuancesShe has noted some differences (e.g., on certain foreign-policy stances such as Iran or Trump), but these are secondary. Her main case is practical delivery, not ideological purity.

@kezia_noble Tried Claude & couldn’t get an answer and apologies not seen your shows . Any key highlights to post here ?

I voted Brexit. I protested against lockdowns. I believe in low tax. I believe in extremely limited immigration. I believe in supporting small businesses. I believe in being tough on crime. I believe in building a better future for our children. Any disagreement with this?

@GriftReport @IfindRetards Surely one for the hall of fame.

Paedophile David Johnson, 35, was caught watching child abuse videos on a train from Stafford to London, with an indecent image set as his phone wallpaper. A horrified passenger spotted him and alerted British Transport Police. He was arrested at Euston station. Johnson had nearly 200 indecent images, including 105 Category A (the most serious). He also pleaded guilty to attempting sexual communication with a child and arranging/facilitating a child sex offence. Jailed for two years and three months at Inner London Crown Court, plus a 10-year Sexual Harm Prevention Order and placed on the sex offenders register for a decade. Just two years is a joke... Sickening. Caught red-handed.

@TheOutlier_1 @claudeai Same experience for me. I'm also on the "Pro" plan. I do quite like Opus 4.6. It's nice for web design. But so is Gemini 3.1 Pro. Claude will absolutely hand their competitors an advantage with these price plans. Someone needs fired.

@GraemeVIP @claudeai Pro. I was about to upgrade to Max a few weeks ago, and luckily they pulled their quota change stunt before I upgraded. Now I just use it for planning to hand off to other agents. Even at 10x or 20x, there's no value there if this is the standard: x.com/TheOutlier_1/s…

@GriftReport He was looking out of sheer disbelief of what he was seeing - some inked up tramp sticking her breakfast in his face for online clout.

OnlyFans girls flashing migrant delivery drivers again and saying he is "disgusting" One day she is going to flash the wrong one..

@TheOutlier_1 @claudeai What plan you on?

@GraemeVIP @claudeai There's zero reason to have multiple sessions with claude anymore. I can burn through my 5 hour window in less than 20minutes on a single session.

"the level of entitlement is insane. did you think anthropic was a charity?" Pathetic strawman fallacy. 🙄 No one asked for charity, you absolute clown. Paying users expected the advertised 5 hour Claude Code limit instead of it getting torched in 8 minutes on normal work while GPT lasts for hours. Anthropic themselves admitted: people are hitting limits "way faster than expected." Your whole post is lazy deflection and idiotic reasoning. That's not debate, that's low IQ reasoning dressed up smug one upmanship.

@GraemeVIP @claudeai the level of entitlement is insane. did you think anthropic was a charity? if you don’t like the pricing, don’t use it.

@isuzixDRnhDW9cF Is Japan going the same way?

英国の現在、日本の未来 🚨🚨🚨📢📢📢 ＞この男は、目の前で女性が頭部に飛来物による致命的な一撃を受けたのを目撃したにもかかわらず、その様子をただ見ているだけだ。しかも……彼の最初の反応は、何事もなかったかのように立ち去ることだった。

@MadalinIgnisca @claudeai It seems flakey. 90% report hit limits too fast. Others like you don't. I wonder why the inconsistency. Maybe someone smarter than me can do a video on this to get to the bottom of it.

@GraemeVIP @claudeai I'm using both Claude and Codex on their maximum plans, and most weeks I don't hit the limit. Rarely maybe 15-25 euros extra in a week when I push a bit more (I enabled extra paid credits both). And thanks to their integration, I use both in Claude Code ;-)

@48statesalbum @AMHeroesUS @claudeai Christ. What is it you're building that takes 8h? How many lines do you get through?

@GraemeVIP @AMHeroesUS @claudeai 20 minutes is far from an autonomous workflow. I regularly tee up tasks (on Claude and codex) that run for upwards of 7/8 hours at a time Works best with opus, or 5.3-codex But 5.4 never keeps enough momentum and always finds itself a reason to stop

@Mindventure_ @claudeai Yes, this is very true. There's a good YouTube video highlighting their money issues.

@GraemeVIP @claudeai Keep in mind, what this means is that OpenAI is just losing a lot more of their investors money on you. At some point, all of this will stop even for Codex. OpenAI is burning money by the boatload.